[SRU Zesty][PATCH 1/1] dma-buf: avoid scheduling on fence status query v2

Alberto Milone alberto.milone at canonical.com
Wed Aug 16 11:59:44 UTC 2017

From: Andres Rodriguez <andresx7 at gmail.com>

When a timeout of zero is specified, the caller is only interested in
the fence status.

In the current implementation, dma_fence_default_wait will always call
schedule_timeout() at least once for an unsignaled fence. This adds a
significant overhead to a fence status query.

Avoid this overhead by returning early if a zero timeout is specified.

v2: move early return after enable_signaling

Signed-off-by: Andres Rodriguez <andresx7 at gmail.com>
Reviewed-by: Christian K├Ânig <christian.koenig at amd.com>
Signed-off-by: Gustavo Padovan <gustavo.padovan at collabora.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170426144620.3560-1-andresx7@gmail.com

BugLink: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1711096

(cherry-picked from commit 03c0c5f6641533f5fc14bf4e76d2304197402552)
Signed-off-by: Alberto Milone <alberto.milone at canonical.com>
 drivers/dma-buf/dma-fence.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/dma-buf/dma-fence.c b/drivers/dma-buf/dma-fence.c
index 0212af7..96b3221 100644
--- a/drivers/dma-buf/dma-fence.c
+++ b/drivers/dma-buf/dma-fence.c
@@ -373,6 +373,11 @@ dma_fence_default_wait(struct dma_fence *fence, bool intr, signed long timeout)
+	if (!timeout) {
+		ret = 0;
+		goto out;
+	}
 	cb.base.func = dma_fence_default_wait_cb;
 	cb.task = current;
 	list_add(&cb.base.node, &fence->cb_list);

More information about the kernel-team mailing list