提交 ad78069c 编写于 作者: C Christian König 提交者: Alex Deucher

drm/amdgpu: fix BO move offsets

It's pretty pointless to get the offset first and then initialize it.

Should fix issues with the new GTT manager.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Reviewed-by: NMichel Dänzer <michel.daenzer@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
上级 98b480db
......@@ -272,8 +272,6 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo,
adev = amdgpu_get_adev(bo->bdev);
ring = adev->mman.buffer_funcs_ring;
old_start = (u64)old_mem->start << PAGE_SHIFT;
new_start = (u64)new_mem->start << PAGE_SHIFT;
switch (old_mem->mem_type) {
case TTM_PL_TT:
......@@ -282,6 +280,7 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo,
return r;
case TTM_PL_VRAM:
old_start = (u64)old_mem->start << PAGE_SHIFT;
old_start += bo->bdev->man[old_mem->mem_type].gpu_offset;
break;
default:
......@@ -295,6 +294,7 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo,
return r;
case TTM_PL_VRAM:
new_start = (u64)new_mem->start << PAGE_SHIFT;
new_start += bo->bdev->man[new_mem->mem_type].gpu_offset;
break;
default:
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册