drm/ttm: fixup ttm_bo_add_move_fence v2
authorMatthew Auld <matthew.auld@intel.com>
Wed, 13 Apr 2022 08:21:33 +0000 (09:21 +0100)
committerChristian König <christian.koenig@amd.com>
Wed, 13 Apr 2022 08:53:09 +0000 (10:53 +0200)
It looks like we still need to call dma_fence_put() on the man->move,
otherwise we just end up leaking it, leading to fireworks later.

v2(Daniel):
  - Simplify the function tail

Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/5689
Fixes: 8bb31587820a ("drm/ttm: remove bo->moving")
Cc: Christian König <christian.koenig@amd.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220413082133.272445-1-matthew.auld@intel.com
drivers/gpu/drm/ttm/ttm_bo.c

index 015a94f766de628b4e117517b643d0eef0baa0cc..75d308ec173d34d0a3e47f5706868e5f491826b9 100644 (file)
@@ -740,11 +740,8 @@ static int ttm_bo_add_move_fence(struct ttm_buffer_object *bo,
        dma_resv_add_fence(bo->base.resv, fence, DMA_RESV_USAGE_KERNEL);
 
        ret = dma_resv_reserve_fences(bo->base.resv, 1);
-       if (unlikely(ret)) {
-               dma_fence_put(fence);
-               return ret;
-       }
-       return 0;
+       dma_fence_put(fence);
+       return ret;
 }
 
 /*