Re: [PATCH] macvtap: use set_curren_state to ensure mb

On Tue, Jun 5, 2012 at 7:05 PM, Eric Dumazet [off-list ref] wrote:

Why not using the more standard :

prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);

and

finish_wait(sk_sleep(sk), &wait);

Thanks for your comments.  The difference is that prepare_to_wait
inside loop introduces extra "list_empty" judgement than original
patch. But personally I prefer the newer version too:

Signed-off-by: Hong Zhiguo <redacted>
---
 drivers/net/macvtap.c |    8 +++-----
 1 files changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
index 2ee56de..0737bd4 100644
--- a/drivers/net/macvtap.c
+++ b/drivers/net/macvtap.c

@@ -847,13 +847,12 @@ static ssize_t macvtap_do_read(struct

macvtap_queue *q, struct kiocb *iocb,
 			       const struct iovec *iv, unsigned long len,
 			       int noblock)
 {
-	DECLARE_WAITQUEUE(wait, current);
+	DEFINE_WAIT(wait);
 	struct sk_buff *skb;
 	ssize_t ret = 0;

-	add_wait_queue(sk_sleep(&q->sk), &wait);
 	while (len) {
-		current->state = TASK_INTERRUPTIBLE;
+		prepare_to_wait(sk_sleep(&q->sk), &wait, TASK_INTERRUPTIBLE);

 		/* Read frames from the queue */
 		skb = skb_dequeue(&q->sk.sk_receive_queue);

@@ -875,8 +874,7 @@ static ssize_t macvtap_do_read(struct

macvtap_queue *q, struct kiocb *iocb,
 		break;
 	}

-	current->state = TASK_RUNNING;
-	remove_wait_queue(sk_sleep(&q->sk), &wait);
+	finish_wait(sk_sleep(&q->sk), &wait);
 	return ret;
 }

-- 
1.7.4.1