struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
struct amdgpu_bo_user *ubo;
- BUG_ON(bo->tbo.type != ttm_bo_type_device);
+ BUG_ON(bo->tbo.type == ttm_bo_type_kernel);
if (adev->family <= AMDGPU_FAMILY_CZ &&
AMDGPU_TILING_GET(tiling_flags, TILE_SPLIT) > 6)
return -EINVAL;
{
struct amdgpu_bo_user *ubo;
- BUG_ON(bo->tbo.type != ttm_bo_type_device);
+ BUG_ON(bo->tbo.type == ttm_bo_type_kernel);
dma_resv_assert_held(bo->tbo.base.resv);
ubo = to_amdgpu_bo_user(bo);
struct amdgpu_bo_user *ubo;
void *buffer;
- BUG_ON(bo->tbo.type != ttm_bo_type_device);
+ BUG_ON(bo->tbo.type == ttm_bo_type_kernel);
ubo = to_amdgpu_bo_user(bo);
if (!metadata_size) {
if (ubo->metadata_size) {
if (!buffer && !metadata_size)
return -EINVAL;
- BUG_ON(bo->tbo.type != ttm_bo_type_device);
+ BUG_ON(bo->tbo.type == ttm_bo_type_kernel);
ubo = to_amdgpu_bo_user(bo);
if (buffer) {
if (buffer_size < ubo->metadata_size)