[PATCH 3/3] drm/radeon: only increment sync_seq when a fence is really emitted

[Nicolai Hähnle]

From: Nicolai Hähnle <nicolai.haeh...@amd.com>

In the rare situation where the kmalloc fails we're probably screwed anyway,
but let's try to be more robust about it.

Signed-off-by: Nicolai Hähnle <Nicolai.Haehnle at amd.com>
---
 drivers/gpu/drm/radeon/radeon_fence.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/radeon/radeon_fence.c 
b/drivers/gpu/drm/radeon/radeon_fence.c
index df09ca7..05815c4 100644
--- a/drivers/gpu/drm/radeon/radeon_fence.c
+++ b/drivers/gpu/drm/radeon/radeon_fence.c
@@ -130,7 +130,7 @@ int radeon_fence_emit(struct radeon_device *rdev,
                      struct radeon_fence **fence,
                      int ring)
 {
-       u64 seq = ++rdev->fence_drv[ring].sync_seq[ring];
+       u64 seq;

        /* we are protected by the ring emission mutex */
        *fence = kmalloc(sizeof(struct radeon_fence), GFP_KERNEL);
@@ -138,7 +138,7 @@ int radeon_fence_emit(struct radeon_device *rdev,
                return -ENOMEM;
        }
        (*fence)->rdev = rdev;
-       (*fence)->seq = seq;
+       (*fence)->seq = seq = ++rdev->fence_drv[ring].sync_seq[ring];
        (*fence)->ring = ring;
        (*fence)->is_vm_update = false;
        fence_init(&(*fence)->base, &radeon_fence_ops,
-- 
2.5.0