linux-stable.git/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c, branch v6.2

drm/amdgpu: Fix potential NULL dereference

2023-01-09T21:49:43+00:00

Fix potential NULL dereference, in the case when "man", the resource manager
might be NULL, when/if we print debug information.

Cc: Alex Deucher 
Cc: Christian König 
Cc: AMD Graphics 
Cc: Dan Carpenter 
Cc: kernel test robot 
Fixes: 7554886daa31ea ("drm/amdgpu: Fix size validation for non-exclusive domains (v4)")
Signed-off-by: Luben Tuikov 
Reviewed-by: Christian König 
Signed-off-by: Alex Deucher

drm/amdgpu: revert "generally allow over-commit during BO allocation"

2022-12-14T21:48:00+00:00

This reverts commit f9d00a4a8dc8fff951c97b3213f90d6bc7a72175.

This causes problem for KFD because when we overcommit we accidentially
bind the BO to GTT for moving it into VRAM. We also need to make sure
that this is done only as fallback after trying to evict first.

Signed-off-by: Christian König 
Reviewed-by: Felix Kuehling 
Signed-off-by: Alex Deucher

drm/amdgpu: Remove unnecessary domain argument

2022-12-14T21:48:00+00:00

Remove the "domain" argument to amdgpu_bo_create_kernel_at() since this
function takes an "offset" argument which is the offset off of VRAM, and as
such allocation always takes place in VRAM. Thus, the "domain" argument is
unnecessary.

Cc: Alex Deucher 
Cc: Christian König 
Cc: AMD Graphics 
Signed-off-by: Luben Tuikov 
Reviewed-by: Christian König 
Signed-off-by: Alex Deucher

drm/amdgpu: Fix size validation for non-exclusive domains (v4)

2022-12-14T21:48:00+00:00

Fix amdgpu_bo_validate_size() to check whether the TTM domain manager for the
requested memory exists, else we get a kernel oops when dereferencing "man".

v2: Make the patch standalone, i.e. not dependent on local patches.
v3: Preserve old behaviour and just check that the manager pointer is not
    NULL.
v4: Complain if GTT domain requested and it is uninitialized--most likely a
    bug.

Cc: Alex Deucher 
Cc: Christian König 
Cc: AMD Graphics 
Signed-off-by: Luben Tuikov 
Reviewed-by: Christian König 
Signed-off-by: Alex Deucher

drm/amdgpu: WARN when freeing kernel memory during suspend

2022-12-14T14:48:04+00:00

When buffers are freed during suspend there is no guarantee that
they can be re-allocated during resume.

The PSP subsystem seems to be quite buggy regarding this, so add
a WARN_ON() to point out those bugs.

Signed-off-by: Christian König 
Reviewed-by: Alex Deucher 
Tested-by: Guilherme G. Piccoli 
Signed-off-by: Alex Deucher

drm/amdgpu: make display pinning more flexible (v2)

2022-12-09T21:50:30+00:00

Only apply the static threshold for Stoney and Carrizo.
This hardware has certain requirements that don't allow
mixing of GTT and VRAM.  Newer asics do not have these
requirements so we should be able to be more flexible
with where buffers end up.

Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/2270
Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/2291
Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/2255
Acked-by: Luben Tuikov 
Reviewed-by: Christian König 
Signed-off-by: Alex Deucher 
Cc: stable@vger.kernel.org

drm/amdgpu: generally allow over-commit during BO allocation

2022-12-06T15:17:24+00:00

We already fallback to a dummy BO with no backing store when we
allocate GDS,GWS and OA resources and to GTT when we allocate VRAM.

Drop all those workarounds and generalize this for GTT as well. This
fixes ENOMEM issues with runaway applications which try to allocate/free
GTT in a loop and are otherwise only limited by the CPU speed.

The CS will wait for the cleanup of freed up BOs to satisfy the
various domain specific limits and so effectively throttle those
buggy applications down to a sane allocation behavior again.

Signed-off-by: Christian König 
Reviewed-by: Alex Deucher 
Reviewed-by: Arunpravin Paneer Selvam 
Signed-off-by: Alex Deucher

drm/ttm: rework on ttm_resource to use size_t type

2022-10-27T09:42:58+00:00

Change ttm_resource structure from num_pages to size_t size in bytes.
v1 -> v2: change PFN_UP(dst_mem->size) to ttm->num_pages
v1 -> v2: change bo->resource->size to bo->base.size at some places
v1 -> v2: remove the local variable
v1 -> v2: cleanup cmp_size_smaller_first()
v2 -> v3: adding missing PFN_UP in ttm_bo_vm_fault_reserved

Signed-off-by: Somalapuram Amaranath 
Link: https://patchwork.freedesktop.org/patch/msgid/20221027091237.983582-1-Amaranath.Somalapuram@amd.com
Reviewed-by: Christian König 
Signed-off-by: Christian König

drm/amdgpu: Set vmbo destroy after pt bo is created

2022-10-06T16:08:09+00:00

Under VRAM usage pression, map to GPU may fail to create pt bo and
vmbo->shadow_list is not initialized, then ttm_bo_release calling
amdgpu_bo_vm_destroy to access vmbo->shadow_list generates below
dmesg and NULL pointer access backtrace:

Set vmbo destroy callback to amdgpu_bo_vm_destroy only after creating pt
bo successfully, otherwise use default callback amdgpu_bo_destroy.

amdgpu: amdgpu_vm_bo_update failed
amdgpu: update_gpuvm_pte() failed
amdgpu: Failed to map bo to gpuvm
amdgpu 0000:43:00.0: amdgpu: Failed to map peer:0000:43:00.0 mem_domain:2
BUG: kernel NULL pointer dereference, address:
 RIP: 0010:amdgpu_bo_vm_destroy+0x4d/0x80 [amdgpu]
 Call Trace:
  
  ttm_bo_release+0x207/0x320 [amdttm]
  amdttm_bo_init_reserved+0x1d6/0x210 [amdttm]
  amdgpu_bo_create+0x1ba/0x520 [amdgpu]
  amdgpu_bo_create_vm+0x3a/0x80 [amdgpu]
  amdgpu_vm_pt_create+0xde/0x270 [amdgpu]
  amdgpu_vm_ptes_update+0x63b/0x710 [amdgpu]
  amdgpu_vm_update_range+0x2e7/0x6e0 [amdgpu]
  amdgpu_vm_bo_update+0x2bd/0x600 [amdgpu]
  update_gpuvm_pte+0x160/0x420 [amdgpu]
  amdgpu_amdkfd_gpuvm_map_memory_to_gpu+0x313/0x1130 [amdgpu]
  kfd_ioctl_map_memory_to_gpu+0x115/0x390 [amdgpu]
  kfd_ioctl+0x24a/0x5b0 [amdgpu]

Signed-off-by: Philip Yang 
Reviewed-by: Christian König 
Signed-off-by: Alex Deucher

Merge drm/drm-next into drm-misc-next

2022-08-01T14:04:00+00:00

Backmerging to pick up fixes from amdgpu.

Signed-off-by: Thomas Zimmermann