Commit 59c66c91 authored by Nicolai Hähnle's avatar Nicolai Hähnle Committed by Alex Deucher

drm/amdgpu: simplify reservation handling during buffer creation

By using ttm_bo_init_reserved instead of the manual initialization of
the reservation object, the reservation lock will be properly unlocked
and destroyed when the TTM BO initialization fails.

Actual deadlocks caused by the missing unlock should have been fixed
by "drm/ttm: never add BO that failed to validate to the LRU list",
superseding the flawed fix in commit 38fc4856 ("drm/amdgpu: fix
a potential deadlock in amdgpu_bo_create_restricted()").

This change fixes remaining recursive locking errors that can be seen
with lock debugging enabled, and avoids the error of freeing a locked
mutex.

As an additional minor bonus, buffers created with resv == NULL and
the AMDGPU_GEM_CREATE_VRAM_CLEARED flag are now only added to the
global LRU list after the fill commands have been issued.

v2: use amdgpu_bo_unreserve instead of ttm_bo_unreserve

Fixes: 12a852219583 ("drm/amdgpu: improve AMDGPU_GEM_CREATE_VRAM_CLEARED handling (v2)")
Signed-off-by: default avatarNicolai Hähnle <nicolai.haehnle@amd.com>
Reviewed-by: default avatarChristian König <christian.koenig@amd.com>
Signed-off-by: default avatarAlex Deucher <alexander.deucher@amd.com>
parent ca9cf68d
...@@ -395,19 +395,10 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev, ...@@ -395,19 +395,10 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev,
amdgpu_fill_placement_to_bo(bo, placement); amdgpu_fill_placement_to_bo(bo, placement);
/* Kernel allocation are uninterruptible */ /* Kernel allocation are uninterruptible */
if (!resv) {
bool locked;
reservation_object_init(&bo->tbo.ttm_resv);
locked = ww_mutex_trylock(&bo->tbo.ttm_resv.lock);
WARN_ON(!locked);
}
initial_bytes_moved = atomic64_read(&adev->num_bytes_moved); initial_bytes_moved = atomic64_read(&adev->num_bytes_moved);
r = ttm_bo_init(&adev->mman.bdev, &bo->tbo, size, type, r = ttm_bo_init_reserved(&adev->mman.bdev, &bo->tbo, size, type,
&bo->placement, page_align, !kernel, NULL, &bo->placement, page_align, !kernel, NULL,
acc_size, sg, resv ? resv : &bo->tbo.ttm_resv, acc_size, sg, resv, &amdgpu_ttm_bo_destroy);
&amdgpu_ttm_bo_destroy);
amdgpu_cs_report_moved_bytes(adev, amdgpu_cs_report_moved_bytes(adev,
atomic64_read(&adev->num_bytes_moved) - initial_bytes_moved); atomic64_read(&adev->num_bytes_moved) - initial_bytes_moved);
...@@ -433,7 +424,7 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev, ...@@ -433,7 +424,7 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev,
dma_fence_put(fence); dma_fence_put(fence);
} }
if (!resv) if (!resv)
ww_mutex_unlock(&bo->tbo.resv->lock); amdgpu_bo_unreserve(bo);
*bo_ptr = bo; *bo_ptr = bo;
trace_amdgpu_bo_create(bo); trace_amdgpu_bo_create(bo);
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment