dma-mapping: fix dma_addressing_limited() if dma_range_map can't cover all system RAM

There is an unusual case that the range map covers right up to the top
of system RAM, but leaves a hole somewhere lower down. Then it prevents
the nvme device dma mapping in the checking path of phys_to_dma() and
causes the hangs at boot.

E.g. On an Armv8 Ampere server, the dsdt ACPI table is:
Method (_DMA, 0, Serialized) // _DMA: Direct Memory Access
{
Name (RBUF, ResourceTemplate ()
{
QWordMemory (ResourceConsumer, PosDecode, MinFixed,
MaxFixed, Cacheable, ReadWrite,
0x0000000000000000, // Granularity
0x0000000000000000, // Range Minimum
0x00000000FFFFFFFF, // Range Maximum
0x0000000000000000, // Translation Offset
0x0000000100000000, // Length
,, , AddressRangeMemory, TypeStatic)
QWordMemory (ResourceConsumer, PosDecode, MinFixed,
MaxFixed, Cacheable, ReadWrite,
0x0000000000000000, // Granularity
0x0000006010200000, // Range Minimum
0x000000602FFFFFFF, // Range Maximum
0x0000000000000000, // Translation Offset
0x000000001FE00000, // Length
,, , AddressRangeMemory, TypeStatic)
QWordMemory (ResourceConsumer, PosDecode, MinFixed,
MaxFixed, Cacheable, ReadWrite,
0x0000000000000000, // Granularity
0x00000060F0000000, // Range Minimum
0x00000060FFFFFFFF, // Range Maximum
0x0000000000000000, // Translation Offset
0x0000000010000000, // Length
,, , AddressRangeMemory, TypeStatic)
QWordMemory (ResourceConsumer, PosDecode, MinFixed,
MaxFixed, Cacheable, ReadWrite,
0x0000000000000000, // Granularity
0x0000007000000000, // Range Minimum
0x000003FFFFFFFFFF, // Range Maximum
0x0000000000000000, // Translation Offset
0x0000039000000000, // Length
,, , AddressRangeMemory, TypeStatic)
})

But the System RAM ranges are:
cat /proc/iomem |grep -i ram
90000000-91ffffff : System RAM
92900000-fffbffff : System RAM
880000000-fffffffff : System RAM
8800000000-bff5990fff : System RAM
bff59d0000-bff5a4ffff : System RAM
bff8000000-bfffffffff : System RAM
So some RAM ranges are out of dma_range_map.

Fix it by checking whether each of the system RAM resources can be
properly encompassed within the dma_range_map.

Signed-off-by: Jia He <justin.he@arm.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: move dma_addressing_limited() out of line

This patch moves dma_addressing_limited() out of line, serving as a
preliminary step to prevent the introduction of a new publicly accessible
low-level helper when validating whether all system RAM is mapped within
the DMA mapping range.

Suggested-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jia He <justin.he@arm.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: move arch_dma_set_mask() declaration to header

This function has a __weak definition and an override that is only used on
freescale powerpc chips. The powerpc definition however does not see the
declaration that is in a .c file:

arch/powerpc/kernel/dma-mask.c:7:6: error: no previous prototype for 'arch_dma_set_mask' [-Werror=missing-prototypes]

Move it into the linux/dma-map-ops.h header where the other arch_dma_* functions
are declared.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: provide CONFIG_ARCH_DMA_DEFAULT_COHERENT

Provide a kconfig option to allow arches to manipulate default
value of dma_default_coherent in Kconfig.

Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: provide a fallback dma_default_coherent

dma_default_coherent was decleared unconditionally at kernel/dma/mapping.c
but only decleared when any of non-coherent options is enabled in
dma-map-ops.h.

Guard the declaration in mapping.c with non-coherent options and provide
a fallback definition.

Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: reject GFP_COMP for noncoherent allocations

While not quite as bogus as for the dma-coherent allocations that were
fixed earlier, GFP_COMP for these allocations has no benefits for
the dma-direct case, and can't be supported at all by dma dma-iommu
backend which splits up allocations into smaller orders. Due to an
oversight in ffcb75458460 that flag stopped being cleared for all
dma allocations, but only got rejected for coherent ones, so fix up
these callers to not allow __GFP_COMP as well after the sound code
has been fixed to not ask for it.

Fixes: ffcb75458460 ("dma-mapping: reject __GFP_COMP in dma_alloc_attrs")
Reported-by: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com>
Reported-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Takashi Iwai <tiwai@suse.de>
Tested-by: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com>
Tested-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>

dma-mapping: reject __GFP_COMP in dma_alloc_attrs

DMA allocations can never be turned back into a page pointer, so
requesting compound pages doesn't make sense and it can't even be
supported at all by various backends.

Reject __GFP_COMP with a warning in dma_alloc_attrs, and stop clearing
the flag in the arm dma ops and dma-iommu.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Marek Szyprowski <m.szyprowski@samsung.com>

dma: kmsan: unpoison DMA mappings

KMSAN doesn't know about DMA memory writes performed by devices. We
unpoison such memory when it's mapped to avoid false positive reports.

Link: https://lkml.kernel.org/r/20220915150417.722975-22-glider@google.com
Signed-off-by: Alexander Potapenko <glider@google.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Andrey Konovalov <andreyknvl@gmail.com>
Cc: Andrey Konovalov <andreyknvl@google.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Christoph Lameter <cl@linux.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Eric Biggers <ebiggers@google.com>
Cc: Eric Biggers <ebiggers@kernel.org>
Cc: Eric Dumazet <edumazet@google.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Ilya Leoshkevich <iii@linux.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Marco Elver <elver@google.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Petr Mladek <pmladek@suse.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Cc: Vegard Nossum <vegard.nossum@oracle.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

dma-mapping: mark dma_supported static

Now that the remaining users in drivers are gone, this function can be
marked static.

Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: reformat comment to suppress htmldoc warning

make html doc reports a cryptic warning with the commit named below:

kernel/dma/mapping.c:258: WARNING: Option list ends without a blank
line; unexpected unindent.

Seems the parser is a bit fussy about the tabbing and having a single
space tab causes the warning. To suppress the warning add another
tab to the list and reindent everything.

Fixes: 7c2645a2a30a ("dma-mapping: allow EREMOTEIO return code for P2PDMA transfers")
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: add flags to dma_map_ops to indicate PCI P2PDMA support

Add a flags member to the dma_map_ops structure with one flag to
indicate support for PCI P2PDMA.

Also, add a helper to check if a device supports PCI P2PDMA.

Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: allow EREMOTEIO return code for P2PDMA transfers

Add EREMOTEIO error return to dma_map_sgtable() which will be used
by .map_sg() implementations that detect P2PDMA pages that the
underlying DMA device cannot access.

Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: add dma_opt_mapping_size()

Streaming DMA mapping involving an IOMMU may be much slower for larger
total mapping size. This is because every IOMMU DMA mapping requires an
IOVA to be allocated and freed. IOVA sizes above a certain limit are not
cached, which can have a big impact on DMA mapping performance.

Provide an API for device drivers to know this "optimal" limit, such that
they may try to produce mapping which don't exceed it.

Signed-off-by: John Garry <john.garry@huawei.com>
Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Acked-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: move pgprot_decrypted out of dma_pgprot

pgprot_decrypted is used by AMD SME systems to allow access to memory
that was set to not encrypted using set_memory_decrypted. That only
happens for dma-direct memory as the IOMMU solves the addressing
challenges for the encryption bit using its own remapping.

Move the pgprot_decrypted call out of dma_pgprot which is also used
by the IOMMU mappings and into dma-direct so that it is only used with
memory that was set decrypted.

Fixes: f5ff79fddf0e ("dma-mapping: remove CONFIG_DMA_REMAP")
Reported-by: Alex Xu (Hello71) <alex_y_xu@yahoo.ca>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Alex Xu (Hello71) <alex_y_xu@yahoo.ca>

ARM: ixp4xx: Drop custom DMA coherency and bouncing

The new PCI driver does not need any of this stuff, so just
drop it.

Cc: iommu@lists.linux-foundation.org
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20220211223238.648934-12-linus.walleij@linaro.org
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>

dma-mapping: remove bogus test for pfn_valid from dma_map_resource

dma_map_resource() uses pfn_valid() to ensure the range is not RAM.
However, pfn_valid() only checks for availability of the memory map for a
PFN but it does not ensure that the PFN is actually backed by RAM.

As dma_map_resource() is the only method in DMA mapping APIs that has this
check, simply drop the pfn_valid() test from dma_map_resource().

Link: https://lore.kernel.org/all/20210824173741.GC623@arm.com/
Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: David Hildenbrand <david@redhat.com>
Link: https://lore.kernel.org/r/20210930013039.11260-2-rppt@kernel.org
Signed-off-by: Will Deacon <will@kernel.org>

dma-debug: teach add_dma_entry() about DMA_ATTR_SKIP_CPU_SYNC

Mapping something twice should be possible as long as,
DMA_ATTR_SKIP_CPU_SYNC is passed to the strictly speaking second relevant
mapping operation (that attempts to map the same thing). So, don't issue a
warning if the specified condition is met in add_dma_entry().

Signed-off-by: Hamza Mahfooz <someguy@effective-light.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: fix the kerneldoc for dma_map_sgtable()

htmldocs began producing the following warnings:

kernel/dma/mapping.c:256: WARNING: Definition list ends without a
blank line; unexpected unindent.
kernel/dma/mapping.c:257: WARNING: Bullet list ends without a blank
line; unexpected unindent.

Reformatting the list without hyphens fixes the warnings and produces
both a readable text and HTML output.

Fixes: fffe3cc8c219 ("dma-mapping: allow map_sg() ops to return negative error code")
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: fix the kerneldoc for dma_map_sg_attrs

Add the missing description for the nents parameter, and fix a trivial
misalignment.

Fixes: fffe3cc8c219 ("dma-mapping: allow map_sg() ops to return negative error codes")
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: return an unsigned int from dma_map_sg{,_attrs}

These can only return 0 for failure or the number of entries, so turn
the return value into an unsigned int.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>

dma-mapping: disallow .map_sg operations from returning zero on error

Now that all the .map_sg operations have been converted to returning
proper error codes, drop the code to handle a zero return value,
add a warning if a zero is returned.

Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: allow map_sg() ops to return negative error codes

Allow dma_map_sgtable() to pass errors from the map_sg() ops. This
will be required for returning appropriate error codes when mapping
P2PDMA memory.

Introduce __dma_map_sg_attrs() which will return the raw error code
from the map_sg operation (whether it be negative or zero). Then add a
dma_map_sg_attrs() wrapper to convert any negative errors to zero to
satisfy the existing calling convention.

dma_map_sgtable() defines three error codes that .map_sg implementations
are allowed to return: -EINVAL, -ENOMEM and -EIO. The latter of which
is a generic return for cases that are passing DMA_MAPPING_ERROR
through.

dma_map_sgtable() will convert a zero error return for old map_sg() ops
into a -EIO return and return any negative errors as reported.

This allows map_sg implementations to start returning multiple
negative error codes. Legacy map_sg implementations can continue
to return zero until they are all converted.

Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: add a dma_alloc_noncontiguous API

Add a new API that returns a potentiall virtually non-contigous sg_table
and a DMA address. This API is only properly implemented for dma-iommu
and will simply return a contigious chunk as a fallback.

The intent is that drivers can use this API if either:

- no kernel mapping or only temporary kernel mappings are required.
That is as a better replacement for DMA_ATTR_NO_KERNEL_MAPPING
- a kernel mapping is required for cached and DMA mapped pages, but
the driver also needs the pages to e.g. map them to userspace.
In that sense it is a replacement for some aspects of the recently
removed and never fully implemented DMA_ATTR_NON_CONSISTENT

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Tomasz Figa <tfiga@chromium.org>
Tested-by: Ricardo Ribalda <ribalda@chromium.org>

dma-mapping: refactor dma_{alloc,free}_pages

Factour out internal versions without the dma_debug calls in preparation
for callers that will need different dma_debug calls.

Note that this changes the dma_debug calls to get the not page aligned
size values, but as long as alloc and free agree on one variant we are
fine.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Tomasz Figa <tfiga@chromium.org>
Tested-by: Ricardo Ribalda <ribalda@chromium.org>

dma-mapping: add a dma_mmap_pages helper

Add a helper to map memory allocated using dma_alloc_pages into
a user address space, similar to the dma_alloc_attrs function for
coherent allocations.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Tomasz Figa <tfiga@chromium.org>
Tested-by: Ricardo Ribalda <ribalda@chromium.org>

dma-mapping: remove the {alloc,free}_noncoherent methods

It turns out allowing non-contigous allocations here was a rather bad
idea, as we'll now need to define ways to get the pages for mmaping
or dma_buf sharing. Revert this change and stick to the original
concept. A different API for the use case of non-contigous allocations
will be added back later.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Tomasz Figa <tfiga@chromium.org>
Tested-by: Ricardo Ribalda <ribalda@chromium.org>:wq

driver core: lift dma_default_coherent into common code

Lift the dma_default_coherent variable from the mips architecture code
to the driver core. This allows an architecture to sdefault all device
to be DMA coherent at run time, even if the kernel is build with support
for DMA noncoherent device. By allowing device_initialize to set the
->dma_coherent field to this default the amount of arch hooks required
for this behavior can be greatly reduced.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de>

dma-mapping: Allow mixing bypass and mapped DMA operation

At the moment we allow bypassing DMA ops only when we can do this for
the entire RAM. However there are configs with mixed type memory
where we could still allow bypassing IOMMU in most cases;
POWERPC with persistent memory is one example.

This adds an arch hook to determine where bypass can still work and
we invoke direct DMA API. The following patch checks the bus limit
on POWERPC to allow or disallow direct mapping.

This adds a ARCH_HAS_DMA_MAP_DIRECT config option to make the arch_xxxx
hooks no-op by default.

Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: move large parts of <linux/dma-direct.h> to kernel/dma

Most of the dma_direct symbols should only be used by direct.c and
mapping.c, so move them to kernel/dma. In fact more of dma-direct.h
should eventually move, but that will require more coordination with
other subsystems.

Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: move dma-debug.h to kernel/dma/

Most of dma-debug.h is not required by anything outside of kernel/dma.
Move the four declarations needed by dma-mappin.h or dma-ops providers
into dma-mapping.h and dma-map-ops.h, and move the remainder of the
file to kernel/dma/debug.h.

Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: split <linux/dma-mapping.h>

Split out all the bits that are purely for dma_map_ops implementations
and related code into a new <linux/dma-map-ops.h> header so that they
don't get pulled into all the drivers. That also means the architecture
specific <asm/dma-mapping.h> is not pulled in by <linux/dma-mapping.h>
any more, which leads to a missing includes that were pulled in by the
x86 or arm versions in a few not overly portable drivers.

Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: add new {alloc,free}_noncoherent dma_map_ops methods

This will allow IOMMU drivers to allocate non-contigous memory and
return a vmapped virtual address.

Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: add a new dma_alloc_pages API

This API is the equivalent of alloc_pages, except that the returned memory
is guaranteed to be DMA addressable by the passed in device. The
implementation will also be used to provide a more sensible replacement
for DMA_ATTR_NON_CONSISTENT flag.

Additionally dma_alloc_noncoherent is switched over to use dma_alloc_pages
as its backend.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> (MIPS part)

dma-mapping: remove dma_cache_sync

All users are gone now, remove the API.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> (MIPS part)

dma-direct: Fix potential NULL pointer dereference

When booting the kernel v5.9-rc4 on a VM, the kernel would panic when
printing a warning message in swiotlb_map(). The dev->dma_mask must not
be a NULL pointer when calling the dma mapping layer. A NULL pointer
check can potentially avoid the panic.

Signed-off-by: Thomas Tai <thomas.tai@oracle.com>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: move dma_common_{mmap,get_sgtable} out of mapping.c

Add a new file that contains helpers for misc DMA ops, which is only
built when CONFIG_DMA_OPS is set.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>

dma-mapping: add a dma_ops_bypass flag to struct device

Several IOMMU drivers have a bypass mode where they can use a direct
mapping if the devices DMA mask is large enough. Add generic support
to the core dma-mapping code to do that to switch those drivers to
a common solution.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>

dma-mapping: move the remaining DMA API calls out of line

For a long time the DMA API has been implemented inline in dma-mapping.h,
but the function bodies can be quite large. Move them all out of line.

This also removes all the dma_direct_* exports as those are just
implementation details and should never be used by drivers directly.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: Alexey Kardashevskiy <aik@ozlabs.ru>

dma-mapping: Add a new dma_need_sync API

Add a new API to check if calls to dma_sync_single_for_{device,cpu} are
required for a given DMA streaming mapping.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20200629130359.2690853-2-hch@lst.de

dma-mapping: Fix dma_pgprot() for unencrypted coherent pages

When dma_mmap_coherent() sets up a mapping to unencrypted coherent memory
under SEV encryption and sometimes under SME encryption, it will actually
set up an encrypted mapping rather than an unencrypted, causing devices
that DMAs from that memory to read encrypted contents. Fix this.

When force_dma_unencrypted() returns true, the linear kernel map of the
coherent pages have had the encryption bit explicitly cleared and the
page content is unencrypted. Make sure that any additional PTEs we set
up to these pages also have the encryption bit cleared by having
dma_pgprot() return a protection with the encryption bit cleared in this
case.

Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Tom Lendacky <thomas.lendacky@amd.com>
Link: https://lkml.kernel.org/r/20200304114527.3636-3-thomas_os@shipmail.org

dma-direct: provide mmap and get_sgtable method overrides

For dma-direct we know that the DMA address is an encoding of the
physical address that we can trivially decode. Use that fact to
provide implementations that do not need the arch_dma_coherent_to_pfn
architecture hook. Note that we still can only support mmap of
non-coherent memory only if the architecture provides a way to set an
uncached bit in the page tables. This must be true for architectures
that use the generic remap helpers, but other architectures can also
manually select it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Max Filippov <jcmvbkbc@gmail.com>

dma-mapping: provide a better default ->get_required_mask

Most dma_map_ops instances are IOMMUs that work perfectly fine in 32-bits
of IOVA space, and the generic direct mapping code already provides its
own routines that is intelligent based on the amount of memory actually
present. Wire up the dma-direct routine for the ARM direct mapping code
as well, and otherwise default to the constant 32-bit mask. This way
we only need to override it for the occasional odd IOMMU that requires
64-bit IOVA support, or IOMMU drivers that are more efficient if they
can fall back to the direct mapping.

Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: remove CONFIG_ARCH_NO_COHERENT_DMA_MMAP

CONFIG_ARCH_NO_COHERENT_DMA_MMAP is now functionally identical to
!CONFIG_MMU, so remove the separate symbol. The only difference is that
arm did not set it for !CONFIG_MMU, but arm uses a separate dma mapping
implementation including its own mmap method, which is handled by moving
the CONFIG_MMU check in dma_can_mmap so that is only applies to the
dma-direct case, just as the other ifdefs for it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> # m68k

dma-mapping: add a dma_can_mmap helper

Add a helper to check if DMA allocations for a specific device can be
mapped to userspace using dma_mmap_*.

Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: explicitly wire up ->mmap and ->get_sgtable

While the default ->mmap and ->get_sgtable implementations work for the
majority of our dma_map_ops impementations they are inherently safe
for others that don't use the page allocator or CMA and/or use their
own way of remapping not covered by the common code. So remove the
defaults if these methods are not wired up, but instead wire up the
default implementations for all safe instances.

Fixes: e1c7e324539a ("dma-mapping: always provide the dma_map_ops based implementation")
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: move the dma_get_sgtable API comments from arm to common code

The comments are spot on and should be near the central API, not just
near a single implementation.

Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: introduce dma_get_merge_boundary()

This patch adds a new DMA API "dma_get_merge_boundary". This function
returns the DMA merge boundary if the DMA layer can merge the segments.
This patch also adds the implementation for a new dma_map_ops pointer.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Reviewed-by: Simon Horman <horms+renesas@verge.net.au>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: remove arch_dma_mmap_pgprot

arch_dma_mmap_pgprot is used for two things:

1) to override the "normal" uncached page attributes for mapping
memory coherent to devices that can't snoop the CPU caches
2) to provide the special DMA_ATTR_WRITE_COMBINE semantics on older
arm systems and some mips platforms

Replace one with the pgprot_dmacoherent macro that is already provided
by arm and much simpler to use, and lift the DMA_ATTR_WRITE_COMBINE
handling to common code with an explicit arch opt-in.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org> # m68k
Acked-by: Paul Burton <paul.burton@mips.com> # mips

dma-mapping: fix page attributes for dma_mmap_*

All the way back to introducing dma_common_mmap we've defaulted to mark
the pages as uncached. But this is wrong for DMA coherent devices.
Later on DMA_ATTR_WRITE_COMBINE also got incorrect treatment as that
flag is only treated special on the alloc side for non-coherent devices.

Introduce a new dma_pgprot helper that deals with the check for coherent
devices so that only the remapping cases ever reach arch_dma_mmap_pgprot
and we thus ensure no aliasing of page attributes happens, which makes
the powerpc version of arch_dma_mmap_pgprot obsolete and simplifies the
remaining ones.

Note that this means arch_dma_mmap_pgprot is a bit misnamed now, but
we'll phase it out soon.

Fixes: 64ccc9c033c6 ("common: dma-mapping: add support for generic dma_mmap_* calls")
Reported-by: Shawn Anastasio <shawn@anastas.io>
Reported-by: Gavin Li <git@thegavinli.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Catalin Marinas <catalin.marinas@arm.com> # arm64

dma-mapping: Remove dma_check_mask()

sme_active() is an x86-specific function so it's better not to call it from
generic code. Christoph Hellwig mentioned that "There is no reason why we
should have a special debug printk just for one specific reason why there
is a requirement for a large DMA mask.", so just remove dma_check_mask().

Signed-off-by: Thiago Jung Bauermann <bauerman@linux.ibm.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20190806044919.10622-4-bauerman@linux.ibm.com

dma-mapping: check pfn validity in dma_common_{mmap,get_sgtable}

Check that the pfn returned from arch_dma_coherent_to_pfn refers to
a valid page and reject the mmap / get_sgtable requests otherwise.

Based on the arm implementation of the mmap and get_sgtable methods.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Vignesh Raghavendra <vigneshr@ti.com>

dma-mapping: truncate dma masks to what dma_addr_t can hold

The dma masks in struct device are always 64-bits wide. But for builds
using a 32-bit dma_addr_t we need to ensure we don't store an
unsupportable value. Before Linux 5.0 this was handled at least by
the ARM dma mapping code by never allowing to set a larger dma_mask,
but these days we allow the driver to just set the largest supported
value and never fall back to a smaller one. Ensure this always works
by truncating the value.

Fixes: 9eb9e96e97b3 ("Documentation/DMA-API-HOWTO: update dma_mask sections")
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: remove an unnecessary NULL check

We already dereferenced "dev" when we called get_dma_ops() so this NULL
check is too late. We're not supposed to pass NULL "dev" pointers to
dma_alloc_attrs().

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>

x86/dma: Remove the x86_dma_fallback_dev hack

Now that we removed support for the NULL device argument in the DMA API,
there is no need to cater for that in the x86 code.

Signed-off-by: Christoph Hellwig <hch@lst.de>

dma: Introduce dma_max_mapping_size()

The function returns the maximum size that can be mapped
using DMA-API functions. The patch also adds the
implementation for direct DMA and a new dma_map_ops pointer
so that other implementations can expose their limit.

Cc: stable@vger.kernel.org
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

dma-mapping, powerpc: simplify the arch dma_set_mask override

Instead of letting the architecture supply all of dma_set_mask just
give it an additional hook selected by Kconfig.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Christian Zigotzky <chzigotzky@xenosoft.de>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

powerpc/dma: stop overriding dma_get_required_mask

The ppc_md and pci_controller_ops methods are unused now and can be
removed. The dma_nommu implementation is generic to the generic one
except for using max_pfn instead of calling into the memblock API,
and all other dma_map_ops instances implement a method of their own.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Christian Zigotzky <chzigotzky@xenosoft.de>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

dma-mapping: remove dmam_{declare,release}_coherent_memory

These functions have never been used.

Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: implement dmam_alloc_coherent using dmam_alloc_attrs

dmam_alloc_coherent is just the default no-flags case of
dmam_alloc_attrs, so take advantage of this similar to the non-managed
version.

Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: fix inverted logic in dma_supported

The cleanup in commit 356da6d0cde3 ("dma-mapping: bypass indirect calls
for dma-direct") accidentally inverted the logic in the check for the
presence of a ->dma_supported() callback. Switch this back to the way it
was to prevent a crash on boot.

Fixes: 356da6d0cde3 ("dma-mapping: bypass indirect calls for dma-direct")
Signed-off-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: bypass indirect calls for dma-direct

Avoid expensive indirect calls in the fast path DMA mapping
operations by directly calling the dma_direct_* ops if we are using
the directly mapped DMA operations.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Tested-by: Jesper Dangaard Brouer <brouer@redhat.com>
Tested-by: Tony Luck <tony.luck@intel.com>

dma-mapping: move dma_cache_sync out of line

This isn't exactly a slow path routine, but it is not super critical
either, and moving it out of line will help to keep the include chain
clean for the following DMA indirection bypass work.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Tested-by: Jesper Dangaard Brouer <brouer@redhat.com>
Tested-by: Tony Luck <tony.luck@intel.com>

dma-mapping: move various slow path functions out of line

There is no need to have all setup and coherent allocation / freeing
routines inline. Move them out of line to keep the implemeation
nicely encapsulated and save some kernel text size.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Tested-by: Jesper Dangaard Brouer <brouer@redhat.com>
Tested-by: Tony Luck <tony.luck@intel.com>

dma-mapping: move dma_get_required_mask to kernel/dma

dma_get_required_mask should really be with the rest of the DMA mapping
implementation instead of in drivers/base as a lone outlier.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Jesper Dangaard Brouer <brouer@redhat.com>
Tested-by: Jesper Dangaard Brouer <brouer@redhat.com>
Tested-by: Tony Luck <tony.luck@intel.com>

dma-mapping: move the remap helpers to a separate file

The dma remap code only makes sense for not cache coherent architectures
(or possibly the corner case of highmem CMA allocations) and currently
is only used by arm, arm64, csky and xtensa. Split it out into a
separate file with a separate Kconfig symbol, which gets the right
copyright notice given that this code was written by Laura Abbott
working for Code Aurora at that point.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Laura Abbott <labbott@redhat.com>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>

dma-mapping: support non-coherent devices in dma_common_get_sgtable

We can use the arch_dma_coherent_to_pfn hook to provide a ->get_sgtable
implementation. Note that this isn't an endorsement of this interface
(which is a horrible bad idea), but it is required to move arm64 over
to the generic code without a loss of functionality.

Signed-off-by: Christoph Hellwig <hch@lst.de>

dma-mapping: consolidate the dma mmap implementations

The only functional differences (modulo a few missing fixes in the arch
code) is that architectures without coherent caches need a hook to
convert a virtual or dma address into a pfn, given that we don't have
the kernel linear mapping available for the otherwise easy virt_to_page
call. As a side effect we can support mmap of the per-device coherent
area even on architectures not providing the callback, and we make
previous dangerous default methods dma_common_mmap actually save for
non-coherent architectures by rejecting it without the right helper.

In addition to that we need a hook so that some architectures can
override the protection bits when mmaping a dma coherent allocations.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Paul Burton <paul.burton@mips.com> # MIPS parts

dma-mapping: remove dma_deconfigure

This goes through a lot of hooks just to call arch_teardown_dma_ops.
Replace it with a direct call instead.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>

dma-mapping: remove dma_configure

There is no good reason for this indirection given that the method
always exists.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>

dma-mapping: move all DMA mapping code to kernel/dma

Currently the code is split over various files with dma- prefixes in the
lib/ and drives/base directories, and the number of files keeps growing.
Move them into a single directory to keep the code together and remove
the file name prefixes. To match the irq infrastructure this directory
is placed under the kernel/ directory.

Signed-off-by: Christoph Hellwig <hch@lst.de>