Am 23.08.2018 um 09:28 schrieb Huang Rui:
> On Wed, Aug 22, 2018 at 05:05:08PM +0200, Christian König wrote:
>> Preparation for following changes. This validates the root PD twice,
>> but the overhead of that should be minimal.
>>
>> Signed-off-by: Christian König <christian.koenig at amd.com>
>> ---
>> drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 8 ++++----
>> 1 file changed, 4 insertions(+), 4 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
>> index 73b8dcaf66e6..53ce9982a5ee 100644
>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
>> @@ -291,11 +291,11 @@ int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm,
>> list_for_each_entry_safe(bo_base, tmp, &vm->evicted, vm_status) {
>> struct amdgpu_bo *bo = bo_base->bo;
>>
>> - if (bo->parent) {
>> - r = validate(param, bo);
>> - if (r)
>> - break;
>> + r = validate(param, bo);
>> + if (r)
>> + break;
> In orignal case, we skip the root PD. But now, it is validated one time.
> May I know where is another time?
That is in the CS code, we add the root PD to the list of BOs which are
validated there.
Christian.
>
> Thanks,
> Ray
>
>>
>> + if (bo->parent) {
>> spin_lock(&glob->lru_lock);
>> ttm_bo_move_to_lru_tail(&bo->tbo);
>> if (bo->shadow)
>> --
>> 2.17.1
>>
>> _______________________________________________
>> amd-gfx mailing list
>> amd-gfx at lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
