drm/amdgpu: add pipeline sync for compute job

hardware ring is async processed, the job is executed in parallel.
In some case, this will result vm fault, like jobs with different vmids.

This works around a CPC hw issue which will eventually be fixed in fw.

Signed-off-by: Chunming Zhou <David1.Zhou@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 692d0d02b644aeac6f5bc78ae4f2a438da8db878..62ce7253e91736ec9fc9267a7e37925e56033d1d 100644 (file)
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -300,7 +300,8 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring,
        int r;
 
        if (ring->funcs->emit_pipeline_sync && (
-           pd_addr != AMDGPU_VM_NO_FLUSH || gds_switch_needed))
+           pd_addr != AMDGPU_VM_NO_FLUSH || gds_switch_needed ||
+                   ring->type == AMDGPU_RING_TYPE_COMPUTE))
                amdgpu_ring_emit_pipeline_sync(ring);
 
        if (ring->funcs->emit_vm_flush &&