[PATCH v2 2/2] drm/amdgpu: expose AMDGPU_GEM_CREATE_VRAM_CLEARED to user space
Christian König
deathsimple at vodafone.de
Wed Aug 3 12:50:23 UTC 2016
Am 01.08.2016 um 15:59 schrieb Alex Deucher:
> On Mon, Aug 1, 2016 at 1:27 AM, Flora Cui <flora.cui at amd.com> wrote:
>> I'll push if no objection
>
> Reviewed-by: Alex Deucher <alexander.deucher at amd.com>
Sorry for the delay, been on vacation last week.
Patch is Reviewed-by: Christian König <christian.koenig at amd.com> as well.
Regards,
Christian.
>
>
>> On Fri, Jul 29, 2016 at 11:23:39AM +0800, Flora Cui wrote:
>>> ping...
>>>
>>> On Mon, Jul 25, 2016 at 02:15:24PM +0800, Flora Cui wrote:
>>>> V2: fix the return value for fill failure and validate bo before
>>>> filling data
>>>>
>>>> Change-Id: I256178afa18c1a433fe60d8656d1c5cc5d55cf2f
>>>> Signed-off-by: Flora Cui <Flora.Cui at amd.com>
>>>> Reviewed-by: Chunming Zhou <David1.Zhou at amd.com>
>>>> ---
>>>> drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 33 ++++++++++++++++++++++++++++++
>>>> include/uapi/drm/amdgpu_drm.h | 2 ++
>>>> 2 files changed, 35 insertions(+)
>>>>
>>>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
>>>> index 6f0873c..59f3a75 100644
>>>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
>>>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
>>>> @@ -277,11 +277,44 @@ int amdgpu_bo_create_restricted(struct amdgpu_device *adev,
>>>> if (unlikely(r != 0)) {
>>>> return r;
>>>> }
>>>> +
>>>> + if (flags & AMDGPU_GEM_CREATE_VRAM_CLEARED &&
>>>> + bo->tbo.mem.placement & TTM_PL_FLAG_VRAM) {
>>>> + struct fence *fence;
>>>> +
>>>> + if (adev->mman.buffer_funcs_ring == NULL ||
>>>> + !adev->mman.buffer_funcs_ring->ready) {
>>>> + r = -EBUSY;
>>>> + goto fail_free;
>>>> + }
>>>> +
>>>> + r = amdgpu_bo_reserve(bo, false);
>>>> + if (unlikely(r != 0))
>>>> + goto fail_free;
>>>> +
>>>> + amdgpu_ttm_placement_from_domain(bo, AMDGPU_GEM_DOMAIN_VRAM);
>>>> + r = ttm_bo_validate(&bo->tbo, &bo->placement, false, false);
>>>> + if (unlikely(r != 0))
>>>> + goto fail_unreserve;
>>>> +
>>>> + amdgpu_fill_buffer(bo, 0, bo->tbo.resv, &fence);
>>>> + amdgpu_bo_fence(bo, fence, false);
>>>> + amdgpu_bo_unreserve(bo);
>>>> + fence_put(bo->tbo.moving);
>>>> + bo->tbo.moving = fence_get(fence);
>>>> + fence_put(fence);
>>>> + }
>>>> *bo_ptr = bo;
>>>>
>>>> trace_amdgpu_bo_create(bo);
>>>>
>>>> return 0;
>>>> +
>>>> +fail_unreserve:
>>>> + amdgpu_bo_unreserve(bo);
>>>> +fail_free:
>>>> + amdgpu_bo_unref(&bo);
>>>> + return r;
>>>> }
>>>>
>>>> int amdgpu_bo_create(struct amdgpu_device *adev,
>>>> diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
>>>> index 946f238..8df3816 100644
>>>> --- a/include/uapi/drm/amdgpu_drm.h
>>>> +++ b/include/uapi/drm/amdgpu_drm.h
>>>> @@ -75,6 +75,8 @@
>>>> #define AMDGPU_GEM_CREATE_NO_CPU_ACCESS (1 << 1)
>>>> /* Flag that USWC attributes should be used for GTT */
>>>> #define AMDGPU_GEM_CREATE_CPU_GTT_USWC (1 << 2)
>>>> +/* Flag that the memory should be in VRAM and cleared */
>>>> +#define AMDGPU_GEM_CREATE_VRAM_CLEARED (1 << 3)
>>>>
>>>> struct drm_amdgpu_gem_create_in {
>>>> /** the requested memory size */
>>>> --
>>>> 1.9.1
>>>>
>>>> _______________________________________________
>>>> amd-gfx mailing list
>>>> amd-gfx at lists.freedesktop.org
>>>> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
>>> _______________________________________________
>>> amd-gfx mailing list
>>> amd-gfx at lists.freedesktop.org
>>> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
>> _______________________________________________
>> amd-gfx mailing list
>> amd-gfx at lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
> _______________________________________________
> amd-gfx mailing list
> amd-gfx at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/amd-gfx
More information about the amd-gfx
mailing list