#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
3c60e59a |
|
03-Jan-2023 |
Rick Chen <rick@andestech.com> |
riscv: memcpy: check src and dst before copy Add src and dst address checking, if they are the same address, just return and don't copy data anymore. Signed-off-by: Rick Chen <rick@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
703b84ec |
|
13-May-2021 |
Bin Meng <bmeng.cn@gmail.com> |
riscv: Fix memmove and optimise memcpy when misalign At present U-Boot SPL fails to boot on SiFive Unleashed board, due to a load address misaligned exception happens when loading the FIT image in spl_load_simple_fit(). The exception happens in memmove() which is called by fdt_splice_(). Commit 8f0dc4cfd106 introduces an assembly version of memmove but it does take misalignment into account (it checks if length is a multiple of machine word size but pointers need also be aligned). As a result it will generate misaligned load/store for the majority of cases and causes significant performance regression on hardware that traps misaligned load/store and emulate them using firmware. The current behaviour of memcpy is that it checks if both src and dest pointers are co-aligned (aka congruent modular SZ_REG). If aligned, it will copy data word-by-word after first aligning pointers to word boundary. If src and dst are not co-aligned, however, byte-wise copy will be performed. This patch was taken from the Linux kernel patch [1], which has not been applied at the time being. It fixes the memmove and optimises memcpy for misaligned cases. It will first align destination pointer to word-boundary regardless whether src and dest are co-aligned or not. If they indeed are, then wordwise copy is performed. If they are not co-aligned, then it will load two adjacent words from src and use shifts to assemble a full machine word. Some additional assembly level micro-optimisation is also performed to ensure more instructions can be compressed (e.g. prefer a0 to t6). With this patch, U-Boot boots again on SiFive Unleashed board. [1] https://patchwork.kernel.org/project/linux-riscv/patch/20210216225555.4976-1-gary@garyguo.net/ Fixes: 8f0dc4cfd106 ("riscv: assembler versions of memcpy, memmove, memset") Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|
#
8f0dc4cf |
|
26-Mar-2021 |
Heinrich Schuchardt <xypron.glpk@gmx.de> |
riscv: assembler versions of memcpy, memmove, memset Provide optimized versions of memcpy(), memmove(), memset() copied from the Linux kernel. Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
|