The root cause of the mentioned test regressions was a sync issue due to
missing/wrong locking of VM status-related fields. It is fixed now, so
the FIXME can go.

Signed-off-by: Natalie Vock <[email protected]>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 20 --------------------
 1 file changed, 20 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c 
b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 6e5db173f6462..4c72162ec331e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -1139,26 +1139,6 @@ static int amdgpu_cs_vm_handling(struct amdgpu_cs_parser 
*p)
                        return r;
        }
 
-       /* FIXME: In theory this loop shouldn't be needed any more when
-        * amdgpu_vm_handle_moved handles all moved BOs that are reserved
-        * with p->ticket. But removing it caused test regressions, so I'm
-        * leaving it here for now.
-        */
-       amdgpu_bo_list_for_each_entry(e, p->bo_list) {
-               bo_va = e->bo_va;
-               if (bo_va == NULL)
-                       continue;
-
-               r = amdgpu_vm_bo_update(adev, bo_va, false);
-               if (r)
-                       return r;
-
-               r = amdgpu_sync_fence(&p->sync, bo_va->last_pt_update,
-                                     GFP_KERNEL);
-               if (r)
-                       return r;
-       }
-
        r = amdgpu_vm_handle_moved(adev, vm, &p->exec.ticket);
        if (r)
                return r;
-- 
2.54.0

Reply via email to