drm/amdgpu: change vm->task_info handling

This patch changes the handling and lifecycle of vm->task_info object.
The major changes are:
- vm->task_info is a dynamically allocated ptr now, and its uasge is
reference counted.
- introducing two new helper funcs for task_info lifecycle management
- amdgpu_vm_get_task_info: reference counts up task_info before
returning this info
- amdgpu_vm_put_task_info: reference counts down task_info
- last put to task_info() frees task_info from the vm.

This patch also does logistical changes required for existing usage
of vm->task_info.

V2: Do not block all the prints when task_info not found (Felix)

V3: Fixed review comments from Felix
- Fix wrong indentation
- No debug message for -ENOMEM
- Add NULL check for task_info
- Do not duplicate the debug messages (ti vs no ti)
- Get first reference of task_info in vm_init(), put last
in vm_fini()

V4: Fixed review comments from Felix
- fix double reference increment in create_task_info
- change amdgpu_vm_get_task_info_pasid
- additional changes in amdgpu_gem.c while porting

Cc: Christian Koenig <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Shashank Sharma <shashank.sharma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: move kiq_reg_write_reg_wait() out of amdgpu_virt.c

It's used for more than just SR-IOV now, so move it to
amdgpu_gmc.c and rename it to better match the functionality and
update the comments in the code paths to better document
when each path is used and why. No functional change.

Reviewed-by: Shaoyun.liu <Shaoyun.liu@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: Shaoyun.Liu@amd.com
Cc: Christian.Koenig@amd.com

drm/amdgpu: Fix ecc irq enable/disable unpaired

The ecc_irq is disabled while GPU mode2 reset suspending process,
but not be enabled during GPU mode2 reset resume process.

Changed from V1:
only do sdma/gfx ras_late_init in aldebaran_mode2_restore_ip
delete amdgpu_ras_late_resume function

Changed from V2:
check umc ras supported before put ecc_irq

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc10: disable AGP aperture

We've had misc reports of random IOMMU page faults when
this is used. It's just a rarely used optimization anyway, so
let's just disable it. It can still be toggled via the
module parameter for testing.

v2: leave it configurable via module parameter

Reviewed-by: Yang Wang <kevinyang.wang@amd.com> (v1)
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Tested-by: Mario Limonciello <mario.limonciello@amd.com> # PHX & Navi33
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add a module parameter to control the AGP aperture

Add a module parameter to control the AGP aperture. The AGP
aperture is an aperture in the GPU's internal address space
which provides direct non-paged access to the platform address
space. This access is non-snooped so only uncached memory
can be accessed.

Add a knob so that we can toggle this for debugging.

Fixes: 67318cb84341 ("drm/amdgpu/gmc11: set gart placement GC11")
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Tested-by: Mario Limonciello <mario.limonciello@amd.com> # PHX & Navi33
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Add xcc param to SRIOV kiq write and WREG32_SOC15_IP_NO_KIQ (v4)

WREG32/RREG32_SOC15_IP_NO_KIQ and amdgpu_virt_kiq_reg_write_reg_wait
are not using the correct rlcg interface or mec engine, respectively.

Add xcc instance parameter to them.

v4: Use GET_INST and squash commit with:
"drm/amdgpu: Add xcc_inst param to amdgpu_virt_kiq_reg_write_reg_wait"

v3: xcc not needed for MMMHUB

v2: rebase

Signed-off-by: Victor Lu <victorchengchi.lu@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix AGP init order

The default AGP settings were overwriting the IP selected
ones since the default was getting set after the IP ones
were selected.

Fixes: de59b69932e6 ("drm/amdgpu/gmc: set a default disable value for AGP")
Link: https://lists.freedesktop.org/archives/amd-gfx/2023-November/100966.html
Tested-by: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com>

drm/amdgpu: cache gpuvm fault information for gmc7+

Cache the current fault info in the vm struct. This can be queried
by userspace later to help debug UMDs.

Cc: samuel.pitoiset@gmail.com
Reviewed-by: Christian König <christian.koenig@amd.com>
Acked-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc: add a way to force a particular placement for GART

We normally place GART based on the location of VRAM and the
available address space around that, but provide an option
to force a particular location for hardware that needs it.

v2: Switch to passing the placement via parameter

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc: set a default disable value for AGP

To disable AGP, the start needs to be set to a higher
value than the end. Set a default disable value for
the AGP aperture and allow the IP specific GMC code
to enable it selectively be calling amdgpu_gmc_agp_location().

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: rework lock handling for flush_tlb v2

Instead of each implementation doing this more or less correctly
move taking the reset lock at a higher level.

v2: fix typo

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: drop error return from flush_gpu_tlb_pasid

That function never fails, drop the error return.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: cleanup gmc_v10_0_flush_gpu_tlb_pasid

The same PASID can be used by more than one VMID, reset each of them.

Use the common KIQ handling.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: rework gmc_v10_0_flush_gpu_tlb v2

Move the SDMA workaround necessary for Navi 1x into a higher layer.

v2: use dev_err

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Add EXT_COHERENT memory allocation flags

These flags (for GEM and SVM allocations) allocate
memory that allows for system-scope atomic semantics.

On GFX943 these flags cause caches to be avoided on
non-local memory.

On all other ASICs they are identical in functionality to the
equivalent COHERENT flags.

Corresponding Thunk patch is at
https://github.com/RadeonOpenCompute/ROCT-Thunk-Interface/pull/88

Reviewed-by: David Yat Sin <David.YatSin@amd.com>
Signed-off-by: David Francis <David.Francis@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Use function for IP version check

Use an inline function for version check. Gives more flexibility to
handle any format changes.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: correct vmhub index in GMC v10/11

Align with new vmhub definition.

v2: use client_id == VMC to decide vmhub(Hawking)

Signed-off-by: Lang Yu <Lang.Yu@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix warnings in gmc_v10_0.c

Fix below checkpatch warnings:

WARNING: Consider removing the code enclosed by this #if 0 and its #endif
WARNING: Prefer 'unsigned int' to bare use of 'unsigned'
WARNING: quoted string split across lines

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Prefer dev_warn over printk

Fix the below warning:

WARNING: Prefer [subsystem eg: netdev]_warn([subsystem]dev, ... then
dev_warn(dev, ... then pr_warn(... to printk(KERN_WARNING ...

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Update invalid PTE flag setting

Update the invalid PTE flag setting with TF enabled.
This is to ensure, in addition to transitioning the
retry fault to a no-retry fault, it also causes the
wavefront to enter the trap handler. With the current
setting, the fault only transitions to a no-retry fault.
Additionally, have 2 sets of invalid PTE settings, one for
TF enabled, the other for TF disabled. The setting with
TF disabled, doesn't work with TF enabled.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Rename DRM schedulers in amdgpu TTM

Rename mman.entity to mman.high_pr to make the distinction
clearer that this is a high priority scheduler. Similarly,
rename the recently added mman.delayed to mman.low_pr to
make it clear it is a low priority scheduler.
No functional change in this patch.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix up missing parameter in kdoc for 'inst' in gmc_ v7, v8, v9, v10, v11.c

Fix these warnings by adding 'inst' arguments to kdocs.

gcc with W=1
drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c:428: warning: Function parameter or member 'inst' not described in 'gmc_v7_0_flush_gpu_tlb_pasid'
drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c:626: warning: Function parameter or member 'inst' not described in 'gmc_v8_0_flush_gpu_tlb_pasid'
drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c:423: warning: Function parameter or member 'inst' not described in 'gmc_v10_0_flush_gpu_tlb_pasid'
drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c:328: warning: Function parameter or member 'inst' not described in 'gmc_v11_0_flush_gpu_tlb_pasid'
drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c:950: warning: Function parameter or member 'inst' not described in 'gmc_v9_0_flush_gpu_tlb_pasid'

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: pass kfd_node ref to svm migration api

This work is required for GC 9.4.3, previous to support memory
partitions per node at SVM. When multiple partition is configured,
every BO should be allocated inside one specific partition which
corresponds to the current amdgpu_device and kfd_node.

v2: squash in compilation fix (Alex)
v3: squash in fix for pre-gfx 9.4.3 (Alex)
v4: squash in best_loc fix (Alex)

Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Add XCC inst to PASID TLB flushing

Add XCC instance to select the correct KIQ ring when
flushing TLBs on a multi-XCC setup.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Tested-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add bitmask to iterate vmhubs

As the layout of VMHUB definition has been changed to cover multiple
XCD/AID case, the original num_vmhubs is not appropriate to do vmhub
iteration any more.

Drop num_vmhubs and introduce vmhubs_mask instead.

v2: switch to the new VMHUB layout
v3: use DECLARE_BITMAP to define vmhubs_mask

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: introduce vmhub definition for multi-partition cases (v3)

v1: Each partition has its own gfxhub or mmhub. adjust
the num of MAX_VMHUBS and the GFXHUB/MMHUB layout (Le)

v2: re-design the AMDGPU_GFXHUB/AMDGPU_MMHUB layout (Le)

v3: apply the gfxhub/mmhub layout to new IPs (Hawking)

v4: fix up gmc11 (Alex)

v5: rebase (Alex)

Signed-off-by: Le Ma <le.ma@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix amdgpu_irq_put call trace in gmc_v10_0_hw_fini

The gmc.ecc_irq is enabled by firmware per IFWI setting,
and the host driver is not privileged to enable/disable
the interrupt. So, it is meaningless to use the amdgpu_irq_put
function in gmc_v10_0_hw_fini, which also leads to the call
trace.

[ 82.340264] Call Trace:
[ 82.340265] <TASK>
[ 82.340269] gmc_v10_0_hw_fini+0x83/0xa0 [amdgpu]
[ 82.340447] gmc_v10_0_suspend+0xe/0x20 [amdgpu]
[ 82.340623] amdgpu_device_ip_suspend_phase2+0x127/0x1c0 [amdgpu]
[ 82.340789] amdgpu_device_ip_suspend+0x3d/0x80 [amdgpu]
[ 82.340955] amdgpu_device_pre_asic_reset+0xdd/0x2b0 [amdgpu]
[ 82.341122] amdgpu_device_gpu_recover.cold+0x4dd/0xbb2 [amdgpu]
[ 82.341359] amdgpu_debugfs_reset_work+0x4c/0x70 [amdgpu]
[ 82.341529] process_one_work+0x21d/0x3f0
[ 82.341535] worker_thread+0x1fa/0x3c0
[ 82.341538] ? process_one_work+0x3f0/0x3f0
[ 82.341540] kthread+0xff/0x130
[ 82.341544] ? kthread_complete_and_exit+0x20/0x20
[ 82.341547] ret_from_fork+0x22/0x30

Signed-off-by: Horatio Zhang <Hongkun.Zhang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix amdgpu_irq_put call trace in gmc_v10_0_hw_fini

The gmc.ecc_irq is enabled by firmware per IFWI setting,
and the host driver is not privileged to enable/disable
the interrupt. So, it is meaningless to use the amdgpu_irq_put
function in gmc_v10_0_hw_fini, which also leads to the call
trace.

[ 82.340264] Call Trace:
[ 82.340265] <TASK>
[ 82.340269] gmc_v10_0_hw_fini+0x83/0xa0 [amdgpu]
[ 82.340447] gmc_v10_0_suspend+0xe/0x20 [amdgpu]
[ 82.340623] amdgpu_device_ip_suspend_phase2+0x127/0x1c0 [amdgpu]
[ 82.340789] amdgpu_device_ip_suspend+0x3d/0x80 [amdgpu]
[ 82.340955] amdgpu_device_pre_asic_reset+0xdd/0x2b0 [amdgpu]
[ 82.341122] amdgpu_device_gpu_recover.cold+0x4dd/0xbb2 [amdgpu]
[ 82.341359] amdgpu_debugfs_reset_work+0x4c/0x70 [amdgpu]
[ 82.341529] process_one_work+0x21d/0x3f0
[ 82.341535] worker_thread+0x1fa/0x3c0
[ 82.341538] ? process_one_work+0x3f0/0x3f0
[ 82.341540] kthread+0xff/0x130
[ 82.341544] ? kthread_complete_and_exit+0x20/0x20
[ 82.341547] ret_from_fork+0x22/0x30

Signed-off-by: Horatio Zhang <Hongkun.Zhang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2522
Fixes: c8b5a95b5709 ("drm/amdgpu: Fix desktop freezed after gpu-reset")
Cc: stable@vger.kernel.org

drm/amdgpu: convert gfx.kiq to array type (v3)

v1: more kiq instances are a available in SOC (Le)
v2: squash commits to avoid breaking the build (Le)
v3: make the conversion for gfx/mec v11_0 (Hawking)

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: move vmhub out of amdgpu_ring_funcs (v4)

It looks better to place this field in ring
structure. Also drop the repeated ring funcs definitions
if there's no difference except for vmhub field.

v2: rename the field to vm_hub like others (Le)
v3: apply the changes to new ip blocks (Hawking)
v4: fix vcn sw ring (Alex)

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Move umc ras block init to gmc ras sw_init

Initialize umc ras block only when umc ip block
supports ras. Driver queries ras capabilities after
early_init, ras block init needs to be moved to
sw_init.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Stanley Yang <Stanley.Yang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add umc retire unit element

It records how many bad pages are retired in one uncorrectable error.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Stanley.Yang <Stanley.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: cleanup visible vram size handling

Centralize the limit handling and validation in one place instead
of spreading that around in different hw generations.

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Luben Tuikov <luben.tuikov@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc10: don't touch gfxhub registers during S0ix

gfxhub registers are part of gfx IP and should not need to be
changed. Doing so without disabling gfxoff can hang the gfx IP.

v2: add comments explaining why we can skip the interrupt
control for S0i3

Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Set MTYPE in PTE based on BO flags

The same BO may need different MTYPEs and SNOOP flags in PTEs depending
on its current location relative to the mapping GPU. Setting MTYPEs from
clients ahead of time is not practical for coherent memory sharing.
Instead determine the correct MTYPE for the desired coherence model and
current BO location when updating the page tables.

To maintain backwards compatibility with MTYPE-selection in
AMDGPU_VA_OP_MAP, the coherence-model-based MTYPE selection is only
applied if it chooses an MTYPE other than MTYPE_NC (the default).

Add two AMDGPU_GEM_CREATE_... flags to indicate the coherence model. The
default if no flag is specified is non-coherent (i.e. coarse-grained
coherent at dispatch boundaries).

Update amdgpu_amdkfd_gpuvm.c to use this new method to choose the
correct MTYPE depending on the current memory location.

v2:
* check that bo is not NULL (e.g. PRT mappings)
* Fix missing ~ bitmask in gmc_v11_0.c
v3:
* squash in "drm/amdgpu: Inherit coherence flags on dmabuf import"

Suggested-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: cleanup scheduler job initialization v2

Init the DRM scheduler base class while allocating the job.

This makes the whole handling much more cleaner.

v2: fix coding style

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20221014084641.128280-7-christian.koenig@amd.com

drm/amdgpu: Increase tlb flush timeout for sriov

[Why]
During multi-vf executing benchmark (Luxmark) observed kiq error timeout.
It happenes because all of VFs do the tlb invalidation at the same time.
Although each VF has the invalidate register set, from hardware side
the invalidate requests are queue to execute.

[How]
In case of 12 VF increase timeout on 12*100ms

Signed-off-by: Dusica Milinkovic <Dusica.Milinkovic@amd.com>
Acked-by: Shaoyun Liu <shaoyun.liu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: enable swiotlb for gmc 10.0 (V2)

Enable swiotlb for gmc 10.0.
v2: include drm_cache.h to use the function ‘drm_need_swiotlb’

Signed-off-by: Aaron Liu <aaron.liu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc10: adjust gart size for parts that support S/G display

For GMC 10 parts which support scatter/gather display (display
from system memory), we should allocate a larger gart size
to better handler larger displays. This mirrors what we already
do for GMC 9 parts.

v2: fix typo (Alex)

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Remove break for VMID loop TLB flush on MES

Loop through all VMIDs for gmc_v10_0_flush_gpu_tlb_pasid and
gmc_v11_0_flush_gpu_tlb_pasid (only if using MES for gmc_v10). This is
required for MES due to use_different_vmid_compute causing SDMA queues
to be assigned different VMIDs than compute for the same PASID.

Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add AMDGPU_VM_NOALLOC v2

Add the AMDGPU_VM_NOALLOC flag to let userspace control MALL allocation.

v2: also add the flag to the allowed flags.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com> (v1)
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: don't use kiq to flush gpu tlb if mes enabled

If MES is enabled, don't use kiq to flush gpu tlb,
for it would result in conflicting with mes fw.

Signed-off-by: Jack Xiao <Jack.Xiao@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc10: skip emitting pasid mapping packet

For MES FW manages IH_VMID_x_LUT updating, skip emitting pasid
mapping packet.

Signed-off-by: Jack Xiao <Jack.Xiao@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: store the mall size in the gmc structure

This will be useful in future patches.

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: keep mmhub clock gating being enabled during s2idle suspend

Without MMHUB clock gating being enabled then MMHUB will not disconnect
from DF and will result in DF C-state entry can't be accessed during S2idle
suspend, and eventually s0ix entry will be blocked.

Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: expand cg_flags from u32 to u64

With this, we can support more CG flags.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: keep mmhub clock gating being enabled during s2idle suspend

Without MMHUB clock gating being enabled then MMHUB will not disconnect
from DF and will result in DF C-state entry can't be accessed during S2idle
suspend, and eventually s0ix entry will be blocked.

Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc: use PCI BARs for APUs in passthrough

If the GPU is passed through to a guest VM, use the PCI
BAR for CPU FB access rather than the physical address of
carve out. The physical address is not valid in a guest.

v2: Fix HDP handing as suggested by Michel

Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Merge get_reserved_allocation to get_vbios_allocations.

Some ASICs need reserved memory for firmware or other components,
which is not allowed to be used by driver.
amdgpu_gmc_get_reserved_allocation is to handle additional areas.
To avoid any missing calling,
merged amdgpu_gmc_get_reserved_allocation to
amdgpu_gmc_get_vbios_allocations.

Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Yongqiang Sun <yongqiang.sun@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Remove redundant .ras_fini initialization in some ras blocks

1. Define amdgpu_ras_block_late_fini_default in amdgpu_ras.c as
.ras_fini common function, which is called when
.ras_fini of ras block isn't initialized.
2. Remove the code of using amdgpu_ras_block_late_fini to
initialize .ras_fini in ras blocks.

Signed-off-by: yipechai <YiPeng.Chai@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Remove redundant calls of amdgpu_ras_block_late_fini in umc ras block

Remove redundant calls of amdgpu_ras_block_late_fini in umc ras block.

Signed-off-by: yipechai <YiPeng.Chai@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add support for gmc10 for gc 10.3.6

this patch adds support for gmc10.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc10: add support for GC 10.3.7

Set gfxhub function and configure VM for GC block.

Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Optimize amdgpu_umc_ras_late_init/amdgpu_umc_ras_fini function code

Optimize amdgpu_umc_ras_late_init/amdgpu_umc_ras_fini function code.

Signed-off-by: yipechai <YiPeng.Chai@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Optimize xxx_ras_late_init/xxx_ras_late_fini for each ras block

1. Define amdgpu_ras_block_late_init to create sysfs nodes
and interrupt handles.
2. Define amdgpu_ras_block_late_fini to remove sysfs nodes
and interrupt handles.
3. Replace ras block variable members in struct
amdgpu_ras_block_object with struct ras_common_if, which
can make it easy to associate each ras block instance
with each ras block functional interface.
4. Add .ras_cb to struct amdgpu_ras_block_object.
5. Change each ras block to fit for the changement of struct
amdgpu_ras_block_object.

Signed-off-by: yipechai <YiPeng.Chai@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add support for GC 10.1.4

Add basic support for GC 10.1.4,
it uses same IP blocks with GC 10.1.3

Signed-off-by: Lang Yu <Lang.Yu@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Move reset sem into reset_domain

We want single instance of reset sem across all
reset clients because in case of XGMI we should stop
access cross device MMIO because any of them could be
in a reset in the moment.

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Link: https://www.spinics.net/lists/amd-gfx/msg74117.html

drm/amdgpu: fix a potential GPU hang on cyan skillfish

We observed a GPU hang when querying GMC CG state(i.e.,
cat amdgpu_pm_info) on cyan skillfish. Acctually, cyan
skillfish doesn't support any CG features.

Just prevent it from accessing GMC CG registers.

Signed-off-by: Lang Yu <Lang.Yu@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Move xgmi ras initialization from .late_init to .early_init

Move xgmi ras initialization from .late_init to .early_init, which let
xgmi ras can be initialized only once.

Signed-off-by: yipechai <YiPeng.Chai@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: remove gart.ready flag

That's just a leftover from old radeon days and was preventing CS and GART
bindings before the hardware was initialized. But nowdays that is
perfectly valid.

The only thing we need to warn about are GART binding before the table
is even allocated.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add vram check function for GMC

This patch will add vram check function for GMC block.
It will write pattern data to the vram and then read back from the vram,
so that to verify the work status of vram.
This patch will cover gmc v6/7/8/9/10.

Signed-off-by: Xiaojian Du <Xiaojian.Du@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix the code style warnings in gmc

Fix the code style warnings in gmc:
ERROR: space required after that ',' (ctx:VxV).

Signed-off-by: yipechai <YiPeng.Chai@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Modify umc block to fit for the unified ras block data and ops

1.Modify umc block to fit for the unified ras block data and ops.
2.Change amdgpu_umc_ras_funcs to amdgpu_umc_ras, and the corresponding variable name remove _funcs suffix.
3.Remove the const flag of umc ras variable so that umc ras block can be able to be inserted into amdgpu device ras block link list.
4.Invoke amdgpu_ras_register_ras_block function to register umc ras block into amdgpu device ras block link list.
5.Remove the redundant code about umc in amdgpu_ras.c after using the unified ras block.
6.Fill unified ras block .name .block .ras_late_init and .ras_fini for all of umc versions. If .ras_late_init and .ras_fini had been defined by the selected umc version, the defined functions will take effect; if not defined, default fill them with amdgpu_umc_ras_late_init and amdgpu_umc_ras_fini.

Signed-off-by: yipechai <YiPeng.Chai@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: John Clements <john.clements@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix a potential GPU hang on cyan skillfish

We observed a GPU hang when querying GMC CG state(i.e.,
cat amdgpu_pm_info) on cyan skillfish. Acctually, cyan
skillfish doesn't support any CG features.

Just prevent it from accessing GMC CG registers.

Signed-off-by: Lang Yu <Lang.Yu@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

drm/amdgpu: recover gart table at resume

Get rid off pin/unpin of gart BO at resume/suspend and
instead pin only once and try to recover gart content
at resume time. This is much more stable in case there
is OOM situation at 2nd call to amdgpu_device_evict_resources()
while evicting GART table.

v3: remove gart recovery from other places
v2: pin gart at amdgpu_gart_table_vram_alloc()

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: get xgmi info before ip_init

Driver needs to call get_xgmi_info() before ip_init
to determine whether it needs to handle a pending hive reset.

Signed-off-by: Victor Skvortsov <victor.skvortsov@amd.com>
Reviewed-by: David Nieto <david.nieto@amd.com>
Reviewed by: shaoyun.liu <Shaoyun.lui@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/amdgpu: fix gmc bo pin count leak in SRIOV

[Why]
gmc bo will be pinned during loading amdgpu and reset in SRIOV while
only unpinned in unload amdgpu

[How]
add amdgpu_in_reset and sriov judgement to skip pin bo

v2: fix wrong judgement

Signed-off-by: Jingwen Chen <Jingwen.Chen2@amd.com>
Reviewed-by: Horace Chen <horace.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: handle IH ring1 overflow

IH ring1 is used to process GPU retry fault, overflow is enabled to
drain retry fault because we want receive other interrupts while
handling retry fault to recover range. There is no overflow flag set
when wptr pass rptr. Use timestamp of rptr and wptr to handle overflow
and drain retry fault.

If fault timestamp goes backward, the fault is filtered and should not
be processed. Drain fault is finished if processed_timestamp is equal to
or larger than checkpoint timestamp.

Add amdgpu_ih_functions interface decode_iv_ts for different chips to
get timestamp from IV entry with different iv size and timestamp offset.
amdgpu_ih_decode_iv_ts_helper is used for vega10, vega20, navi10.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: convert IP version array to include instances

Allow us to query instances versions more cleanly.

Instancing support is not consistent unfortunately. SDMA is a
good example. Sienna cichlid has 4 total SDMA instances, each
enumerated separately (HWIDs 42, 43, 68, 69). Arcturus has 8
total SDMA instances, but they are enumerated as multiple
instances of the same HWIDs (4x HWID 42, 4x HWID 43). UMC
is another example. On most chips there are multiple
instances with the same HWID. This allows us to support both
forms.

v2: rebase
v3: clarify instancing support

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc10.0: convert to IP version checking

Use IP versions rather than asic_type to differentiate
IP version specific features.

v2: squash in gmc fixes
v3: rebase

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix gart.bo pin_count leak

gmc_v{9,10}_0_gart_disable() isn't called matched with
correspoding gart_enbale function in SRIOV case. This will
lead to gart.bo pin_count leak on driver unload.

Cc: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Leslie Shi <Yuliang.Shi@amd.com>
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix gart.bo pin_count leak

gmc_v{9,10}_0_gart_disable() isn't called matched with
correspoding gart_enbale function in SRIOV case. This will
lead to gart.bo pin_count leak on driver unload.

Cc: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Leslie Shi <Yuliang.Shi@amd.com>
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: check access permisson to restore retry fault

Check range access permission to restore GPU retry fault, if GPU retry
fault on address which belongs to VMA, and VMA has no read or write
permission requested by GPU, failed to restore the address. The vm fault
event will pass back to user space.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add cyan_skillfish support in gmc v10

Add gmc support for cyan_skillfish.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add gpu harvest support for yellow carp (v2)

Register callback in gfxhub functions to program the bypass groups in
gc_utcl2 corresponding to harvested SA.

v2: update comments (Alex)

Signed-off-by: Xiaomeng Hou <Xiaomeng.Hou@amd.com>
Signed-off-by: Aaron Liu <aaron.liu@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: introduce a stolen reserved buffer to protect specific buffer region (v2)

Some ASICs such as Yellow Carp needs to reserve a region of video memory
to avoid access from driver. So this patch is to introduce a stolen
reserved buffer to protect specific buffer region.

v2: free this buffer in amdgpu_ttm_fini.

Signed-off-by: Huang Rui <ray.huang@amd.com>
Acked-and-Tested-by: Aaron Liu <aaron.liu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add gmc v10 supports for yellow carp

Add gfx memory controller support for yellow carp.

Signed-off-by: Aaron Liu <aaron.liu@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/amdgpu/gmc_v10_0: Fix potential copy/paste issue

Fixes the following W=1 kernel build warning(s):

drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c:955: warning: expecting prototype for gmc_v8_0_gart_fini(). Prototype was for gmc_v10_0_gart_fini() instead

Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Christian König" <christian.koenig@amd.com>
Cc: David Airlie <airlied@linux.ie>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Cc: amd-gfx@lists.freedesktop.org
Cc: dri-devel@lists.freedesktop.org
Cc: linux-media@vger.kernel.org
Cc: linaro-mm-sig@lists.linaro.org
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Modify GC register access from MMIO to RLCG in file amdgpu_gmc.c

In SRIOV environment, KMD should access GC registers
with RLCG if GC indirect access flag enabled.

Signed-off-by: Peng Ju Zhou <PengJu.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Handle IOMMU enabled case.

Problem:
Handle all DMA IOMMU group related dependencies before the
group is removed. Those manifest themself in that when IOMMU
enabled DMA map/unmap is dependent on the presence of IOMMU
group the device belongs to but, this group is released once
the device is removed from PCI topology.

Fix:
Expedite all such unmap operations to pci remove driver callback.

v5: Drop IOMMU notifier and switch to lockless call to ttm_tt_unpopulate
v6: Drop the BO unamp list
v7:
Drop amdgpu_gart_fini
In amdgpu_ih_ring_fini do uncinditional check (!ih->ring)
to avoid freeing uniniitalized rings.
Call amdgpu_ih_ring_fini unconditionally.
v8: Add deatiled explanation

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210517143851.475058-1-andrey.grodzovsky@amd.com

drm/amd/amdgpu: add gmc ip block for beige_goby

Enable gmc block for beige_goby, same as sienna_cichlid

Signed-off-by: Chengming Gui <Jack.Gui@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/amdgpu: add gmc support for beige_goby

Same as dimgrey_cavefish

Signed-off-by: Chengming Gui <Jack.Gui@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Refine the error report when flush tlb.

there are 2 hubs to flush in the gmc, to make it easier
to debug when hub flush failed, refine the logs.

Signed-off-by: Peng Ju Zhou <PengJu.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Calling address translation functions to simplify codes

Use amdgpu_gmc_vram_pa and amdgpu_gmc_vram_cpu_pa
to simplify codes. No logic change.

Signed-off-by: Oak Zeng <Oak.Zeng@amd.com>
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: split umc callbacks to ras and non-ras ones

umc ras is not managed by gpu driver when gpu is
connected to cpu through xgmi. split umc callbacks
into ras and non-ras ones so gpu driver only
initializes umc ras callbacks when it manages
umc ras.

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Dennis Li <Dennis.Li@amd.com>
Reviewed-by: John Clements <John.Clements@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Replace in_interrupt() usage in gmc_v*_process_interrupt()

The usage of in_interrupt() in gmc_v*_process_interrupt() is intended to
use a different code path if invoked from the interrupt handler vs
invoked from the workqueue.

The usage of in_interrupt() in drivers is phased out and Linus clearly
requested that code which changes behaviour depending on context should
either be separated or the context be conveyed in an argument passed by the
caller, which usually knows the context.

gmc_v*_process_interrupt() is invoked via the ->process() callback
from amdgpu_ih_process() which in turn is invoked either from
amdgpu_irq_handler() (the interrupt handler) or from
amdgpu_irq_handle_*() which is a workqueue.

amdgpu_irq::ih is always processed from the interrupt handler, the other
three struct amdgpu_ih_ring members are processed from a workqueue.

Replace the in_interrupt() check with a comparison against adev->irq.ih.
A similar check is already done to check if the ih pointer is from
ih_soft.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Print the IH client ID name when vm fault happens

This gives more information and improves productivity.

Signed-off-by: Yong Zhao <Yong.Zhao@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: switch hdp callback functions for hdp v5

Switch to use the HDP functions which unified on hdp structure instead of
the scattered hdp callback functions.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: remove amdgpu_ttm_late_init and amdgpu_bo_late_init

No longer used.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: use AMDGPU_NUM_VMID when possible

Replace hardcoded vmid number with AMDGPU_NUM_VMID macro.

Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/amdgpu/gmc_v10_0: Suppy some missing function doc descriptions

Fixes the following W=1 kernel build warning(s):

drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c:278: warning: Function parameter or member 'vmhub' not described in 'gmc_v10_0_flush_gpu_tlb'
drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c:278: warning: Function parameter or member 'flush_type' not described in 'gmc_v10_0_flush_gpu_tlb'
drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c:371: warning: Function parameter or member 'flush_type' not described in 'gmc_v10_0_flush_gpu_tlb_pasid'
drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c:371: warning: Function parameter or member 'all_hub' not described in 'gmc_v10_0_flush_gpu_tlb_pasid'

Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Christian König" <christian.koenig@amd.com>
Cc: David Airlie <airlied@linux.ie>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Sumit Semwal <sumit.semwal@linaro.org>
Cc: amd-gfx@lists.freedesktop.org
Cc: dri-devel@lists.freedesktop.org
Cc: linux-media@vger.kernel.org
Cc: linaro-mm-sig@lists.linaro.org
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: enable AGP aperture on gmc10.x (v2)

Just a small optimization for accessing system pages directly.
Was missed for gmc v10 since the feature landed for older gmcs
while we were still on the emulator or gmc10 and we use the AGP
aperture for zfb on the emulator.

v2: fix up the system aperture as well

Reviewed-and-tested-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: implement retry fault handling for Navi

Same as gmc9, basically filter the fault, reroute or handle it.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: cleanup gmc_v10_0_process_interrupt a bit

Return early in case of a ratelimit and don't print leading zeros for
the address.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: improve code indentation and alignment

General code indentation and alignment changes such as replace spaces
by tabs or align function arguments as per the coding style
guidelines. The patch covers various .c files for this driver.
Issue reported by checkpatch script.

Signed-off-by: Deepak R Varma <mh12gx2825@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add support to configure MALL for sienna_cichlid (v2)

Enable Memory Access at Last Level (MALL) feature for sienna_cichlid.

v2: drop module option. We need to add UAPI so userspace can
request MALL per buffer.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc10: remove dummy read workaround for newer chips

Sienna Cichlid and newer have a hw fix so no longer require
the workaround.

Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add gmc cg support for dimgrey_cavefish

The athub version for dimgrey_cavefish is v2.1.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add gmc ip block for dimgrey_cavefish

Enable gmc block for dimgrey_cavefish, same as sienna_cichlid.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Jiansong Chen <Jiansong.Chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add gmc support for dimgrey_cavefish

Same as navy_flounder.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Jiansong Chen <Jiansong.Chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add mmhub v2.3 for vangogh (v4)

There are too many register offset mismatch between mmhub v2.0 and v2.3.

E.X:

mmMM_ATC_L2_MISC_CG: 0x064a(v2.0) 0x06cd(v2.3)
mmMMVM_L2_PROTECTION_FAULT_CNTL: 0x0688(v2.0) 0x0708(v2.3)
mmMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32: 0x072b(v2.0) 0x0940(v2.3)
mmMMVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32: 0x072c(v2.0) 0x0941(v2.3)
mmMMVM_INVALIDATE_ENG0_REQ: 0x06e3(v2.0) 0x0a01(v2.3)
mmMMVM_INVALIDATE_ENG0_ACK: 0x06f5(v2.0) 0x0a02(v2.3)
mmMMVM_CONTEXT0_CNTL: 0x06c0(v2.0) 0x0740(v2.3)
mmMMVM_L2_PROTECTION_FAULT_STATUS: 0x068c(v2.0) 0x070c(v2.3)
mmMMVM_L2_PROTECTION_FAULT_CNTL: 0x0688(v2.0) 0x0708(v2.3)
mmMM_ATC_L2_MISC_CG: 0x064a(v2.0) 0x06cd(v2.3)
mmDAGB0_CNTL_MISC2: 0x0071(v2.0) 0x0096(v2.3)
...

Continuing using the same file mmhub v2.0 is not good choice, it will
introduce a lot of checking with ASIC types. And also easy to introduce the
issues that offset not align, this kind of issues are really hard to find. Van
Gogh's mmhub vm invalidation is actually caused by the offset mismatch as well.

So it would like to create a new file rather than stick to re-use orignal mmhub
v2.0 here.

v2: add missed translate_further programming.
v3: sync with latest code
v4: add missing callbacks

Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add gmc v10 supports for van gogh (v4)

Add gfx memory controller support for van gogh.

v2: don't use dynamic invalidate eng allocation for van gogh.
v3: squash in other fixes
v4: rebase

Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: use function pointer for gfxhub functions

gfxhub functions are now called from function pointers,
instead of from asic-specific functions.

Signed-off-by: Oak Zeng <Oak.Zeng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: block ring buffer access during GPU recovery

When GPU is in reset, its status isn't stable and ring buffer also need
be reset when resuming. Therefore driver should protect GPU recovery
thread from ring buffer accessed by other threads. Otherwise GPU will
randomly hang during recovery.

v2: correct indent

Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: refine message print for devices of hive

Using dev_xxx instead of DRM_xxx/pr_xxx to indicate which device
of a hive is the message for.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: refine codes to avoid reentering GPU recovery

if other threads have holden the reset lock, recovery will
fail to try_lock. Therefore we introduce atomic hive->in_reset
and adev->in_gpu_reset, to avoid reentering GPU recovery.

v2:
drop "? true : false" in the definition of amdgpu_in_reset

Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: revert "fix system hang issue during GPU reset"

The whole approach wasn't thought through till the end.

We already had a reset lock like this in the past and it caused the same problems like this one.

Completely revert the patch for now and add individual trylock protection to the hardware access functions as necessary.

This reverts commit df9c8d1aa278c435c30a69b8f2418b4a52fcb929.

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Use function pointer for some mmhub functions

Add more function pointers to amdgpu_mmhub_funcs. ASIC specific
implementation of most mmhub functions are called from a general
function pointer, instead of calling different function for
different ASIC. Simplify the code by deleting duplicate functions

Signed-off-by: Oak Zeng <Oak.Zeng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc10: switch to using amdgpu_gmc_get_vbios_allocations

The new helper centralizes the logic in one place.

Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: move stolen vga bo from amdgpu to amdgpu.gmc

Since that is where we store the other data related to
the stolen vga memory.

Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: use a define for the memory size of the vga emulator

Rather than open coding it everywhere.

Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: enable RAS support for sienna cichlid

enabled GECC error injection and query support

Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: John Clements <john.clements@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: enable umc 8.7 functions in gmc v10

add support for umc 8.7 initialization

add umc 8.7 source to makefile

Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: John Clements <john.clements@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix system hang issue during GPU reset

when GPU hang, driver has multi-paths to enter amdgpu_device_gpu_recover,
the atomic adev->in_gpu_reset and hive->in_reset are used to avoid
re-entering GPU recovery.

During GPU reset and resume, it is unsafe that other threads access GPU,
which maybe cause GPU reset failed. Therefore the new rw_semaphore
adev->reset_sem is introduced, which protect GPU from being accessed by
external threads during recovery.

v2:
1. add rwlock for some ioctls, debugfs and file-close function.
2. change to use dqm->is_resetting and dqm_lock for protection in kfd
driver.
3. remove try_lock and change adev->in_gpu_reset as atomic, to avoid
re-enter GPU recovery for the same GPU hang.

v3:
1. change back to use adev->reset_sem to protect kfd callback
functions, because dqm_lock couldn't protect all codes, for example:
free_mqd must be called outside of dqm_lock;

[ 1230.176199] Hardware name: Supermicro SYS-7049GP-TRT/X11DPG-QT, BIOS 3.1 05/23/2019
[ 1230.177221] Call Trace:
[ 1230.178249] dump_stack+0x98/0xd5
[ 1230.179443] amdgpu_virt_kiq_reg_write_reg_wait+0x181/0x190 [amdgpu]
[ 1230.180673] gmc_v9_0_flush_gpu_tlb+0xcc/0x310 [amdgpu]
[ 1230.181882] amdgpu_gart_unbind+0xa9/0xe0 [amdgpu]
[ 1230.183098] amdgpu_ttm_backend_unbind+0x46/0x180 [amdgpu]
[ 1230.184239] ? ttm_bo_put+0x171/0x5f0 [ttm]
[ 1230.185394] ttm_tt_unbind+0x21/0x40 [ttm]
[ 1230.186558] ttm_tt_destroy.part.12+0x12/0x60 [ttm]
[ 1230.187707] ttm_tt_destroy+0x13/0x20 [ttm]
[ 1230.188832] ttm_bo_cleanup_memtype_use+0x36/0x80 [ttm]
[ 1230.189979] ttm_bo_put+0x1be/0x5f0 [ttm]
[ 1230.191230] amdgpu_bo_unref+0x1e/0x30 [amdgpu]
[ 1230.192522] amdgpu_amdkfd_free_gtt_mem+0xaf/0x140 [amdgpu]
[ 1230.193833] free_mqd+0x25/0x40 [amdgpu]
[ 1230.195143] destroy_queue_cpsch+0x1a7/0x270 [amdgpu]
[ 1230.196475] pqm_destroy_queue+0x105/0x260 [amdgpu]
[ 1230.197819] kfd_ioctl_destroy_queue+0x37/0x70 [amdgpu]
[ 1230.199154] kfd_ioctl+0x277/0x500 [amdgpu]
[ 1230.200458] ? kfd_ioctl_get_clock_counters+0x60/0x60 [amdgpu]
[ 1230.201656] ? tomoyo_file_ioctl+0x19/0x20
[ 1230.202831] ksys_ioctl+0x98/0xb0
[ 1230.204004] __x64_sys_ioctl+0x1a/0x20
[ 1230.205174] do_syscall_64+0x5f/0x250
[ 1230.206339] entry_SYSCALL_64_after_hwframe+0x49/0xbe

2. remove try_lock and introduce atomic hive->in_reset, to avoid
re-enter GPU recovery.

v4:
1. remove an unnecessary whitespace change in kfd_chardev.c
2. remove comment codes in amdgpu_device.c
3. add more detailed comment in commit message
4. define a wrap function amdgpu_in_reset

v5:
1. Fix some style issues.

Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Suggested-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Suggested-by: Christian König <christian.koenig@amd.com>
Suggested-by: Felix Kuehling <Felix.Kuehling@amd.com>
Suggested-by: Lijo Lazar <Lijo.Lazar@amd.com>
Suggested-by: Luben Tukov <luben.tuikov@amd.com>
Signed-off-by: Dennis Li <Dennis.Li@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: won't include gc and mmhub register headers in GMC block

All gc/mmhub register access and operation should be in gfxhub/mmhub level.

Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: move get_invalidate_req function into gfxhub/mmhub level

This patch is to move get_invalidate_req into gfxhub/mmhub level. It will avoid
mismatch of the different gfxhub/mmhub register offsets and fields in the same
gmc block.

Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add vmhub funcs helper (v2)

This patch is to introduce vmhub funcs helper to add following callback
(print_l2_protection_fault_status). Each GC/MMHUB register specific programming
should be in gfxhub/mmhub level.

v2: remove the condition of funcs assignment.

Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: abstract set_vm_fault_masks function to refine the programming

This patch is to add set_vm_fault_masks helper to amdgpu_gmc to refine the
original programming.

Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add gmc cg support for navy_flounder

The athub version used for navy_flounder is v2.1.

Signed-off-by: Jiansong Chen <Jiansong.Chen@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add gmc ip block for navy_flounder

navy_flounder has similar gc IP version with sienna_cichlid,
follow its setting for the moment.

Signed-off-by: Jiansong Chen <Jiansong.Chen@amd.com>
Reviewed-by: Tao Zhou <Tao.Zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc10: add navy_flounder support

Same as navi10.

Signed-off-by: Jiansong Chen <Jiansong.Chen@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: use register distance member instead of hardcode in GMC10

This patch updates to use register distance member instead of hardcode
in GMC10.

Signed-off-by: Huang Rui <ray.huang@amd.com>
Tested-by: AnZhong Huang <anzhong.huang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Clean up KFD VMID assignment

The KFD VMID assignment was hard-coded in a few places. Consolidate that in
a single variable adev->vm_manager.first_kfd_vmid. The value is still
assigned in gmc-ip-version-specific code.

Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add XGMI support for sienna cichlid

support for setting up XGMI FB address regions

Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: John Clements <john.clements@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/sriov : Use kiq to do tlb invalidation for gfx10 on sriov

On SRIOV run time, driver shouldn't directly access invalidation registers through MMIO.
Use kiq to submit wait_reg_mem package for the invalidation

Signed-off-by: shaoyunl <shaoyun.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add gmc cg support for sienna_cichlid

Add gmc clockgating support for sienna_cichlid.
The athub version used for sienna_cichlid is v2.1.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add gmc ip block for sienna_cichlid

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc10: add sienna_cichlid support

Same as navi10.

Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: resize VRAM BAR for CPU access on gfx10

Try to resize BAR0 to let CPU access all of VRAM on Navi. Syncs
code with previous gfx generations from commit d6895ad39f3b39
("drm/amdgpu: resize VRAM BAR for CPU access v6").

Signed-off-by: Alan Swanson <reiver@improbability.net>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: cleanup IB pool handling a bit

Fix the coding style, move and rename the definitions to
better match what they are supposed to be doing.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc: Use consistent variable on unlocks

Currently the error returns paths are unlocking lock kiq->ring_lock
however it seems this should be dev->gfx.kiq.ring_lock as this
is the lock that is being locked and unlocked around the ring
operations. This looks like a bug, but it's not. The kiq is just
a local variable pointing to the same structure. Make it consistent.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: protect ring overrun

Wait for the oldest sequence on the ring
to be signaled in order to make sure there
will be no command overrun.

v2: fix coding stype and remove abs operation
v3: remove the initialization of variable r

Signed-off-by: Yintian Tao <yttao@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Print UTCL2 client ID on a gpuvm fault

UTCL2 client ID is useful information to get which
UTCL2 client caused the gpuvm fault. Print it out
for debug purpose

Signed-off-by: Oak Zeng <Oak.Zeng@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian Konig <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: implement more ib pools (v2)

We have three ib pools, they are normal, VM, direct pools.

Any jobs which schedule IBs without dependence on gpu scheduler should
use DIRECT pool.

Any jobs schedule direct VM update IBs should use VM pool.

Any other jobs use NORMAL pool.

v2: squash in coding style fix

Signed-off-by: xinhui pan <xinhui.pan@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Use the correct flush_type in flush_gpu_tlb_pasid

The flush_type was incorrectly hard-coded to 0 when calling falling back
to MMIO-based invalidation in flush_gpu_tlb_pasid.

Fixes: ea930000a6dc ("drm/amdgpu: export function to flush TLB via pasid")
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Oak Zeng <Oak.Zeng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix TLB invalidation request when using semaphore

Use a more meaningful variable name for the invalidation request
that is distinct from the tmp variable that gets overwritten when
acquiring the invalidation semaphore.

Fixes: 4ed8a03740d0 ("drm/amdgpu: invalidate mmhub semaphore workaround in gmc9/gmc10")
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Yong Zhao <Yong.Zhao@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Revert "drm/amdgpu: fix modprobe failure of the secondary GPU when GDDR6 training enabled(V5)"

This reverts commit 9e441478623fd913d4340654682b19f0c24e629d.

The patch will be replaced with a better solution, revert it.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Tianci.Yin <tianci.yin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: modify packet size for pm4 flush tlbs

[Why]
PM4 packet size for flush message was oversized.

[How]
Packet size adjusted to allocate flush + fence packets.

Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix modprobe failure of the secondary GPU when GDDR6 training enabled(V5)

[why]
In dual GPUs scenario, stolen_size is assigned to zero on the secondary GPU,
since there is no pre-OS console using that memory. Then the bottom region of
VRAM was allocated as GTT, unfortunately a small region of bottom VRAM was
encroached by UMC firmware during GDDR6 BIST training, this cause page fault.

[how]
Forcing stolen_size to 3MB, then the bottom region of VRAM was
allocated as stolen memory, GTT corruption avoid.

Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Signed-off-by: Tianci.Yin <tianci.yin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: export function to flush TLB via pasid

This can be used directly from amdgpu and amdkfd to invalidate
TLB through pasid.
It supports gmc v7, v8, v9 and v10.

Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc10: free stolen memory in late_init

We don't need to store the pre-OS console memory after
the driver has loaded so free it.

Reviewed-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc10: remove dead code

Leftover from bring up. We look up the actual pre-OS memory usage
value later in the same function.

Reviewed-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: catch amdgpu_irq_add_id failure

Do not ignore amdgpu_irq_add_id return value while registering
VMC page fault interrupt.

Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc10: use common invalidation engine helper

Rather than open coding it. This also changes the free masks
to better reflect the usage by other components.

Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add invalidate semaphore limit for SRIOV in gmc10

It may fail to load guest driver in round 2 when using invalidate
semaphore for SRIOV. So it needs to avoid using invalidate semaphore
for SRIOV.

Signed-off-by: changzhu <Changfeng.Zhu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: use CPU to flush vmhub if sched stopped

otherwse the flush_gpu_tlb will hang if we unload the
KMD becuase the schedulers already stopped

Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: use CPU to flush vmhub if sched stopped

otherwse the flush_gpu_tlb will hang if we unload the
KMD becuase the schedulers already stopped

Signed-off-by: Monk Liu <Monk.Liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: invalidate mmhub semaphore workaround in gmc9/gmc10

It may lose gpuvm invalidate acknowldege state across power-gating off
cycle. To avoid this issue in gmc9/gmc10 invalidation, add semaphore acquire
before invalidation and semaphore release after invalidation.

After adding semaphore acquire before invalidation, the semaphore
register become read-only if another process try to acquire semaphore.
Then it will not be able to release this semaphore. Then it may cause
deadlock problem. If this deadlock problem happens, it needs a semaphore
firmware fix.

Signed-off-by: changzhu <Changfeng.Zhu@amd.com>
Acked-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

drm/amdgpu: invalidate mmhub semaphore workaround in gmc9/gmc10

It may lose gpuvm invalidate acknowldege state across power-gating off
cycle. To avoid this issue in gmc9/gmc10 invalidation, add semaphore acquire
before invalidation and semaphore release after invalidation.

After adding semaphore acquire before invalidation, the semaphore
register become read-only if another process try to acquire semaphore.
Then it will not be able to release this semaphore. Then it may cause
deadlock problem. If this deadlock problem happens, it needs a semaphore
firmware fix.

Signed-off-by: changzhu <Changfeng.Zhu@amd.com>
Acked-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add dummy read by engines for some GCVM status registers in gfx10

The GRBM register interface is now capable of bursting 1 cycle per
register wr->wr, wr->rd much faster than previous muticycle per
transaction done interface. This has caused a problem where
status registers requiring HW to update have a 1 cycle delay, due
to the register update having to go through GRBM.

For cp ucode, it has realized dummy read in cp firmware.It covers
the use of WAIT_REG_MEM operation 1 case only.So it needs to call
gfx_v10_0_wait_reg_mem in gfx10. Besides it also needs to add warning to
update firmware in case firmware is too old to have function to realize
dummy read in cp firmware.

For sdma ucode, it hasn't realized dummy read in sdma firmware. sdma is
moved to gfxhub in gfx10. So it needs to add dummy read in driver
between amdgpu_ring_emit_wreg and amdgpu_ring_emit_reg_wait for sdma_v5_0.

Signed-off-by: changzhu <Changfeng.Zhu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add dummy read by engines for some GCVM status registers in gfx10

The GRBM register interface is now capable of bursting 1 cycle per
register wr->wr, wr->rd much faster than previous muticycle per
transaction done interface. This has caused a problem where
status registers requiring HW to update have a 1 cycle delay, due
to the register update having to go through GRBM.

For cp ucode, it has realized dummy read in cp firmware.It covers
the use of WAIT_REG_MEM operation 1 case only.So it needs to call
gfx_v10_0_wait_reg_mem in gfx10. Besides it also needs to add warning to
update firmware in case firmware is too old to have function to realize
dummy read in cp firmware.

For sdma ucode, it hasn't realized dummy read in sdma firmware. sdma is
moved to gfxhub in gfx10. So it needs to add dummy read in driver
between amdgpu_ring_emit_wreg and amdgpu_ring_emit_reg_wait for sdma_v5_0.

Signed-off-by: changzhu <Changfeng.Zhu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/sdma5: do not execute 0-sized IBs (v2)

This seems to help with https://bugs.freedesktop.org/show_bug.cgi?id=111481.

v2: insert a NOP instead of skipping all 0-sized IBs to avoid breaking older hw

Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/sdma5: do not execute 0-sized IBs (v2)

This seems to help with https://bugs.freedesktop.org/show_bug.cgi?id=111481.

v2: insert a NOP instead of skipping all 0-sized IBs to avoid breaking older hw

Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Report vram vendor with sysfs (v3)

The vram vendor can be found as a separate sysfs file at:
/sys/class/drm/card[X]/device/mem_info_vram_vendor
The vram vendor is displayed as a string value.

v2: Use correct bit masking, and cache vram_vendor in gmc
v3: Drop unused functions for vram width, type, and vendor

Signed-off-by: Ori Messinger <ori.messinger@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/atomfirmware: simplify the interface to get vram info

fetch both the vram type and width in one function call. This
avoids having to parse the same data table twice to get the two
pieces of data.

Reviewed-by: Xiaojie Yuan <xiaojie.yuan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/SRIOV: Navi12 SRIOV VF gets GTT base

With changes in PSP and HV, SRIOV VF will handle

vram gtt location just like bare metal. There is

no need to differentiate it anymore.

Signed-off-by: Jiange Zhao <Jiange.Zhao@amd.com>
Reviewed-by: Emily Deng <Emily.Deng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: cleanup PTE flag generation v3

Move the ASIC specific code into a new callback function.

v2: mask the flags for SI and CIK instead of a BUG_ON().
v3: remove last missed BUG_ON().

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: cleanup mtype mapping

Unify how we map the UAPI flags to the PTE hardware flags for a mapping.

Only the MTYPE is actually ASIC dependent, all other flags should be
copied over 1 to 1 and ASIC differences are handled later on.

Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: switch to new amdgpu_nbio structure

no functional change, just switch to new structures

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add dummy read for some GCVM status registers

The GRBM register interface is now capable of bursting 1 cycle per
register wr->wr, wr->rd much faster than previous muticycle per
transaction done interface. This has caused a problem where status
registers requiring HW to update have a 1 cycle delay, due to the
register update having to go through GRBM.

SW may operate on an incorrect value if they write a register and
immediately check the corresponding status register.

Registers requiring HW to clear or set fields may be delayed by 1 cycle.
For example,

1. write VM_INVALIDATE_ENG0_REQ mask = 5a
2. read VM_INVALIDATE_ENG0_ACK till the ack is same as the request mask = 5a
a. HW will reset VM_INVALIDATE_ENG0_ACK = 0 until invalidation is complete
3. write VM_INVALIDATE_ENG0_REQ mask = 5a
4. read VM_INVALIDATE_ENG0_ACK till the ack is same as the request mask = 5a
a. First read of VM_INVALIDATE_ENG0_ACK = 5a instead of 0
b. Second read of VM_INVALIDATE_ENG0_ACK = 0 because
the remote GRBM h/w register takes one extra cycle to be cleared
c. In this case, SW will see a false ACK if they exit on first read

Affected registers (only GC variant) | Recommended Dummy Read
--------------------------------------+----------------------------
VM_INVALIDATE_ENG*_ACK | VM_INVALIDATE_ENG*_REQ
VM_L2_STATUS | VM_L2_STATUS
VM_L2_PROTECTION_FAULT_STATUS | VM_L2_PROTECTION_FAULT_STATUS
VM_L2_PROTECTION_FAULT_ADDR_HI/LO32 | VM_L2_PROTECTION_FAULT_ADDR_HI/LO32
VM_L2_IH_LOG_BUSY | VM_L2_IH_LOG_BUSY
MC_VM_L2_PERFCOUNTER_HI/LO | MC_VM_L2_PERFCOUNTER_HI/LO
ATC_L2_PERFCOUNTER_HI/LO | ATC_L2_PERFCOUNTER_HI/LO
ATC_L2_PERFCOUNTER2_HI/LO | ATC_L2_PERFCOUNTER2_HI/LO

Signed-off-by: Xiaojie Yuan <xiaojie.yuan@amd.com>
Reviewed-by: Jack Xiao <Jack.Xiao@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Add more page fault info printing for GFX10

The printing we did for GFX9 was not propogated to GFX10 somehow, so fix
it now.

Signed-off-by: Yong Zhao <Yong.Zhao@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Export function to flush TLB of specific vm hub

This is for kfd to reuse amdgpu TLB invalidation function.
On gfx10, kfd only needs to flush TLB on gfx hub but not
on mm hub. So export a function for KFD flush TLB only on
specific hub.

Signed-off-by: Oak Zeng <Oak.Zeng@amd.com>
Reviewed-by: Christian Konig <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: simplify and cleanup setting the dma mask

Use dma_set_mask_and_coherent to set both masks in one go, and remove
the no longer required fallback, as the kernel now always accepts
larger than required DMA masks. Fail the driver probe if we can't
set the DMA mask, as that means the system can only support a larger
mask.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc10: set gart size and vm size for navi12

Same as other navi asics.

Signed-off-by: Xiaojie Yuan <xiaojie.yuan@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add new member in amdgpu_device for vmhub counts per asic chip

It aims to replace AMDGPU_MAX_VMHUBS in for loop to initialize registers.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: rename AMDGPU_GFXHUB/MMHUB macro with hub number

The number of GFXHUB/MMHUB may be expanded in later ASICs.

Signed-off-by: Le Ma <le.ma@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gmc10: add navi14 support

same as navi10

Signed-off-by: Xiaojie Yuan <xiaojie.yuan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jack Xiao <Jack.Xiao@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/powerplay: add baco smu reset function for smu11

add baco reset support for smu11.
it can help gpu do asic reset when gpu recovery.

Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add gmc v10 ip block for navi10 (v6)

GMC in the GPU memory controller.

v1: add place holder and initial basic implementation (Ray)
v2: retire unused amdgpu_gart_set_defaults (Hawking)
v3: re-work get_vm_pde function (Hawking)
v4: replace legacy amdgpu_vram/gtt_location with
amdgpu_gmc_vram/gtt_location (Hawking)
v5: squash in updates (Alex)
v6: use get_vbios_fb_size (Alex)

Signed-off-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>