drm/amdgpu: create GDS, GWS and OA in system domain

And only move them in on validation. This allows for better control
when multiple processes are fighting over those resources.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Tested-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This commit is contained in:
Christian König 2019-05-16 10:25:35 +02:00 committed by Alex Deucher
parent a3e7738d57
commit 94de7349f7

View File

@ -495,7 +495,11 @@ static int amdgpu_bo_do_create(struct amdgpu_device *adev,
#endif
bo->tbo.bdev = &adev->mman.bdev;
amdgpu_bo_placement_from_domain(bo, bp->domain);
if (bp->domain & (AMDGPU_GEM_DOMAIN_GWS | AMDGPU_GEM_DOMAIN_OA |
AMDGPU_GEM_DOMAIN_GDS))
amdgpu_bo_placement_from_domain(bo, AMDGPU_GEM_DOMAIN_CPU);
else
amdgpu_bo_placement_from_domain(bo, bp->domain);
if (bp->type == ttm_bo_type_kernel)
bo->tbo.priority = 1;