On 10/6/2014 4:22 PM, Michael S. Tsirkin wrote:

quoted

quoted

quoted

quoted

config_mutex served two purposes: prevent multiple concurrent config
change handlers, and synchronize access to config_enable flag.

quoted

quoted

quoted

quoted

Since commit dbf2576e37da0fcc7aacbfbb9fd5d3de7888a3c1
     workqueue: make all workqueues non-reentrant
all workqueues are non-reentrant, and config_enable
is now gone.

quoted

quoted

quoted

quoted

Get rid of the unnecessary lock.

quoted

quoted

quoted

quoted

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
  drivers/net/virtio_net.c | 7 +------
  1 file changed, 1 insertion(+), 6 deletions(-)

quoted

quoted

quoted

quoted

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index fa17afa..d80fef4 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c

[...]

quoted

@@ -1430,7 +1426,7 @@ static void virtnet_config_changed_work(struct work_struct *work)
  		netif_tx_stop_all_queues(vi->dev);
  	}
  done:
-	mutex_unlock(&vi->config_lock);
+	return;

quoted

quoted

quoted

    There's no need for this *return*.

quoted

quoted

I know - it's removed by the follow-up patch.

quoted

    Yeah, I saw.

quoted

quoted

It's formatted like this to make diff smaller
and make review easier.

quoted

    Don't understand how adding this line makes diff smaller though...
You first need to add it and then to delete it, where's the save?

quoted

WBR, Sergei

If I don't add it, gcc generates a compiler warning: it does not like
labels at the end of functions.

    Ahh... nevermind then.

WBR, Sergei