lineage-rockchip/android_external_minigbm
1
0
Fork 0
Code Activity Actions

Revert "Revert "amdgpu: Set USWC for SCANOUT images.""

Browse source 
This reverts commit 91d36976aa84abde87f0f56d235d57042ca3ea81.i

The previous changes should fix the mapping performance with USWC
memory.

original change:
    amdgpu: Set USWC for SCANOUT images.

    This is necessary for scanout from GTT. Without it, the kernel will
    migrate it to the carveout. For large resolutions this would not be
    feasible due to small carveout sizes and hence it is important that
    we alloc scanout from GTT.

    The original patch was on top of
    4eebcea "amdgpu: make AMDGPU_GEM_CREATE_CPU_GTT_USWC flag default"
    which set USWC unconditionally, but has since been reverted. The
    main failure reason is being addressed in mesa-amd. This patch also
    still disables USWC if we can and the image will be read often on
    the CPU, to avoid staging textures during memory mapping.

TEST=Run android.hardware.camera2.cts.RecordingTest#testVideoPreviewSurfaceSharing on Grunt.
BUG=b:152378755

Change-Id: Ice0041de6676c057a2659986b962c2d54833e928
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/2246025
Tested-by: Bas Nieuwenhuizen <basni@chromium.org>
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Commit-Queue: Bas Nieuwenhuizen <basni@chromium.org>
This commit is contained in:
Bas Nieuwenhuizen 2020-06-04 23:11:27 +02:00 • committed by Commit Bot
parent 4a3f98cc34
commit 4daf12cff0
1 changed files with 5 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

6
amdgpu.c
View file

@ -426,7 +426,11 @@ static int amdgpu_create_bo_linear(struct bo *bo, uint32_t width, uint32_t heigh
gem_create.in.domain_flags |= AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
gem_create.in.domains = AMDGPU_GEM_DOMAIN_GTT;
if (!(use_flags & (BO_USE_SW_READ_OFTEN | BO_USE_SCANOUT)))
/* Scanout in GTT requires USWC, otherwise try to use cachable memory
* for buffers that are read often, because uncacheable reads can be
* very slow. USWC should be faster on the GPU though. */
if ((use_flags & BO_USE_SCANOUT) || !(use_flags & BO_USE_SW_READ_OFTEN))
gem_create.in.domain_flags |= AMDGPU_GEM_CREATE_CPU_GTT_USWC;
/* Allocate the buffer with the preferred heap. */