[net-next rfc V3 2/9] macvtap: do not add self to waitqueue if doing a nonblock read

[Jason Wang]

There's no need to add self to waitqueue if doing a nonblock read. This could
help to avoid the spinlock contention.

Signed-off-by: Jason Wang <jasowang@xxxxxxxxxx>
---
 drivers/net/macvtap.c |    7 +++++--
 1 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
index 5e485e3..8949631 100644
--- a/drivers/net/macvtap.c
+++ b/drivers/net/macvtap.c
@@ -845,7 +845,9 @@ static ssize_t macvtap_do_read(struct macvtap_queue *q, struct kiocb *iocb,
 	ssize_t ret = 0;
 
 	while (len) {
-		prepare_to_wait(sk_sleep(&q->sk), &wait, TASK_INTERRUPTIBLE);
+		if (!noblock)
+			prepare_to_wait(sk_sleep(&q->sk), &wait,
+					TASK_INTERRUPTIBLE);
 
 		/* Read frames from the queue */
 		skb = skb_dequeue(&q->sk.sk_receive_queue);
@@ -867,7 +869,8 @@ static ssize_t macvtap_do_read(struct macvtap_queue *q, struct kiocb *iocb,
 		break;
 	}
 
-	finish_wait(sk_sleep(&q->sk), &wait);
+	if (!noblock)
+		finish_wait(sk_sleep(&q->sk), &wait);
 	return ret;
 }
 
-- 
1.7.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/