drm/amdgpu: don't finish the ring if not initialized
authorTrigger Huang <trigger.huang@amd.com>
Tue, 8 Aug 2017 10:42:51 +0000 (06:42 -0400)
committerAlex Deucher <alexander.deucher@amd.com>
Tue, 15 Aug 2017 18:46:17 +0000 (14:46 -0400)
If a ring is not initialized, it also should not be finished.
For example, in Vega10's SR-IOV environment, UVD's decode ring is not
initialized, but will be finnished in amdgpu_uvd_sw_fini, because UVD
driver put all the uvd decode ring's finish operation into
amdgpu_uvd_sw_fini function, while not uvd_vXXX_0_sw_fini. This will
lead to amdgpu module unloading failure.

Signed-off-by: Trigger Huang <trigger.huang@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c

index 70447567438a06d6b723d2c4d25680c7529c46de..6c5646b48d1a5fce145df441ce66098ea62e95d8 100644 (file)
@@ -261,6 +261,10 @@ void amdgpu_ring_fini(struct amdgpu_ring *ring)
 {
        ring->ready = false;
 
+       /* Not to finish a ring which is not initialized */
+       if (!(ring->adev) || !(ring->adev->rings[ring->idx]))
+               return;
+
        amdgpu_wb_free(ring->adev, ring->rptr_offs);
        amdgpu_wb_free(ring->adev, ring->wptr_offs);