On Mon, Apr 14, 2025 at 9:26 AM Sasha Levin <sas...@kernel.org> wrote:
>
> From: Christian König <christian.koe...@amd.com>
>
> [ Upstream commit f5e7fabd1f5c65b2e077efcdb118cfa67eae7311 ]
>
> Try pinning into VRAM to allow P2P with RDMA NICs without ODP
> support if all attachments can do P2P. If any attachment can't do
> P2P just pin into GTT instead.
>
> Acked-by: Simona Vetter <simona.vet...@ffwll.ch>
> Signed-off-by: Christian König <christian.koe...@amd.com>
> Signed-off-by: Felix Kuehling <felix.kuehl...@amd.com>
> Reviewed-by: Felix Kuehling <felix.kuehl...@amd.com>
> Tested-by: Pak Nin Lui <pak....@amd.com>
> Cc: Simona Vetter <simona.vet...@ffwll.ch>
> Signed-off-by: Alex Deucher <alexander.deuc...@amd.com>
> Signed-off-by: Sasha Levin <sas...@kernel.org>
This should not go to stable. It depends on dmem cgroups.
Alex
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c | 25 +++++++++++++++------
> 1 file changed, 18 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
> b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
> index 9f627caedc3f6..ee47f239c8c12 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
> @@ -75,11 +75,25 @@ static int amdgpu_dma_buf_attach(struct dma_buf *dmabuf,
> */
> static int amdgpu_dma_buf_pin(struct dma_buf_attachment *attach)
> {
> - struct drm_gem_object *obj = attach->dmabuf->priv;
> - struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
> + struct dma_buf *dmabuf = attach->dmabuf;
> + struct amdgpu_bo *bo = gem_to_amdgpu_bo(dmabuf->priv);
> + u32 domains = bo->preferred_domains;
>
> - /* pin buffer into GTT */
> - return amdgpu_bo_pin(bo, AMDGPU_GEM_DOMAIN_GTT);
> + dma_resv_assert_held(dmabuf->resv);
> +
> + /*
> + * Try pinning into VRAM to allow P2P with RDMA NICs without ODP
> + * support if all attachments can do P2P. If any attachment can't do
> + * P2P just pin into GTT instead.
> + */
> + list_for_each_entry(attach, &dmabuf->attachments, node)
> + if (!attach->peer2peer)
> + domains &= ~AMDGPU_GEM_DOMAIN_VRAM;
> +
> + if (domains & AMDGPU_GEM_DOMAIN_VRAM)
> + bo->flags |= AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
> +
> + return amdgpu_bo_pin(bo, domains);
> }
>
> /**
> @@ -134,9 +148,6 @@ static struct sg_table *amdgpu_dma_buf_map(struct
> dma_buf_attachment *attach,
> r = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
> if (r)
> return ERR_PTR(r);
> -
> - } else if (bo->tbo.resource->mem_type != TTM_PL_TT) {
> - return ERR_PTR(-EBUSY);
> }
>
> switch (bo->tbo.resource->mem_type) {
> --
> 2.39.5
>
