virtio-net: don't respond to cpu hotplug notifier if we're not ready
authorJason Wang <jasowang@redhat.com>
Tue, 15 Oct 2013 03:18:58 +0000 (11:18 +0800)
committerDavid S. Miller <davem@davemloft.net>
Thu, 17 Oct 2013 19:53:09 +0000 (15:53 -0400)
We're trying to re-configure the affinity unconditionally in cpu hotplug
callback. This may lead the issue during resuming from s3/s4 since

- virt queues haven't been allocated at that time.
- it's unnecessary since thaw method will re-configure the affinity.

Fix this issue by checking the config_enable and do nothing is we're not ready.

The bug were introduced by commit 8de4b2f3ae90c8fc0f17eeaab87d5a951b66ee17
(virtio-net: reset virtqueue affinity when doing cpu hotplug).

Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
drivers/net/virtio_net.c

index defec2b3c5a408ff035889d015d717f040e2864c..c4bc1cc7ef337d32af5ef4feb1bbb88ebacee083 100644 (file)
@@ -1116,6 +1116,11 @@ static int virtnet_cpu_callback(struct notifier_block *nfb,
 {
        struct virtnet_info *vi = container_of(nfb, struct virtnet_info, nb);
 
+       mutex_lock(&vi->config_lock);
+
+       if (!vi->config_enable)
+               goto done;
+
        switch(action & ~CPU_TASKS_FROZEN) {
        case CPU_ONLINE:
        case CPU_DOWN_FAILED:
@@ -1128,6 +1133,9 @@ static int virtnet_cpu_callback(struct notifier_block *nfb,
        default:
                break;
        }
+
+done:
+       mutex_unlock(&vi->config_lock);
        return NOTIFY_OK;
 }