drm/amdgpu: set the contiguous flag if possible

This avoids reallocating scanout BOs on first pin in a lot of cases.

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Tested-by: Nirmoy Das <nirmoy.das@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index e2cbe19404c0e6c5e72abe63ad17e157521e6dae..87c40c4cdd50a2742f1038c71b364dab9f38face 100644 (file)
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -446,10 +446,11 @@ static int amdgpu_vram_mgr_new(struct ttm_resource_manager *man,
        }
        spin_unlock(&mgr->lock);
 
-       atomic64_add(vis_usage, &mgr->vis_usage);
+       if (i == 1)
+               mem->placement |= TTM_PL_FLAG_CONTIGUOUS;
 
+       atomic64_add(vis_usage, &mgr->vis_usage);
        mem->mm_node = nodes;
-
        return 0;
 
 error: