When TTM failed to alloc VRAM, TTM evict BOs from VRAM to system memory
then retry the allocation, this currently skips the KFD BOs from the
same process because KFD requires all BOs are resident for user queues.
If TTM BO with TTM_PL_FLAG_CONTIGUOUS flag to alloc contiguous VRAM,
allow TTM evict KFD BOs from the same process, this will evict the user
queues first, and restore the queues later after contiguous VRAM
allocation.
Signed-off-by: Philip Yang <philip.y...@amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index a5ceec7820cf..00b8603d73e5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1383,7 +1383,8 @@ static bool amdgpu_ttm_bo_eviction_valuable(struct
ttm_buffer_object *bo,
*/
dma_resv_for_each_fence(&resv_cursor, bo->base.resv,
DMA_RESV_USAGE_BOOKKEEP, f) {
- if (amdkfd_fence_check_mm(f, current->mm))
+ if (amdkfd_fence_check_mm(f, current->mm) &&
+ !(place->flags & TTM_PL_FLAG_CONTIGUOUS))
return false;
}
--
2.43.2
