From: Herbert Xu <herbert.xu@redhat.com>
Date: Thu, 7 Jan 2010 12:41:39 -0500
Subject: [net] virtio_net: fix tx wakeup race condition
Message-id: <20100107124139.GA10078@gondor.apana.org.au>
Patchwork-id: 22333
O-Subject: [RHEL5.5 PATCH] virtio_net: Fix tx wakeup race condition
Bugzilla: 524651
RH-Acked-by: Mark McLoughlin <markmc@redhat.com>
RH-Acked-by: David S. Miller <davem@redhat.com>
RH-Acked-by: Avi Kivity <avi@redhat.com>

Hi:

RHEL5 Bugzilla #524651

This patch fixes a race condition that causes TX to completely
stop in a guest using virtio_net.  This bug doesn't exist upstream
as virtio_net has diverged from the RHEL5.4 code base.

This fix has been verified by the bug reporter.

virtio_net: Fix tx wakeup race condition

We free completed TX requests in xmit_tasklet but do not wake
the queue.  This creates a race condition whereupon the queue
may be emptied by xmit_tasklet and yet it remains in the stopped
state.

This patch fixes this by waking the queue after freeing packets
in xmit_tasklet.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

Cheers,

Signed-off-by: Jarod Wilson <jarod@redhat.com>

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 9eec5a5..74b3854 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -521,8 +521,10 @@ static void xmit_tasklet(unsigned long data)
 		vi->svq->vq_ops->kick(vi->svq);
 		vi->last_xmit_skb = NULL;
 	}
-	if (vi->free_in_tasklet)
+	if (vi->free_in_tasklet) {
 		free_old_xmit_skbs(vi);
+		netif_wake_queue(vi->dev);
+	}
 	netif_tx_unlock_bh(vi->dev);
 }