drm/amdgpu: use LRU mapping policy for SDMA engines (90c11309) · Commits · 戴 / test

Spreading the load across multiple SDMA engines can increase memory transfer performance. Signed-off-by:

drivers/gpu/drm/amd/amdgpu/amdgpu_queue_mgr.c

+1 −1

Original line number	Diff line number	Diff line
		@@ -276,7 +276,6 @@ int amdgpu_queue_mgr_map(struct amdgpu_device *adev,

		switch (mapper->hw_ip) {
		case AMDGPU_HW_IP_GFX:
		case AMDGPU_HW_IP_DMA:
		case AMDGPU_HW_IP_UVD:
		case AMDGPU_HW_IP_VCE:
		case AMDGPU_HW_IP_UVD_ENC:
		@@ -284,6 +283,7 @@ int amdgpu_queue_mgr_map(struct amdgpu_device *adev,
		case AMDGPU_HW_IP_VCN_ENC:
		r = amdgpu_identity_map(adev, mapper, ring, out_ring);
		break;
		case AMDGPU_HW_IP_DMA:
		case AMDGPU_HW_IP_COMPUTE:
		r = amdgpu_lru_map(adev, mapper, ring, out_ring);
		break;