vhost: use mutex_lock_nested() in vhost_dev_lock_vqs()

jasowang · gregkh · commit bd3ccdc6f922 · 2018-02-25T11:05:39.000+01:00
commit e9cb423 upstream. We used to call mutex_lock() in vhost_dev_lock_vqs() which tries to hold mutexes of all virtqueues. This may confuse lockdep to report a possible deadlock because of trying to hold locks belong to same class. Switch to use mutex_lock_nested() to avoid false positive. Fixes: 6b1e6cc ("vhost: new device IOTLB API") Reported-by: syzbot+dbb7c1161485e61b0241@syzkaller.appspotmail.com Signed-off-by: Jason Wang <jasowang@redhat.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
@@ -849,7 +849,7 @@ static void vhost_dev_lock_vqs(struct vhost_dev *d)
 {
 	int i = 0;
 	for (i = 0; i < d->nvqs; ++i)
-		mutex_lock(&d->vqs[i]->mutex);
+		mutex_lock_nested(&d->vqs[i]->mutex, i);
 }
 
 static void vhost_dev_unlock_vqs(struct vhost_dev *d)

Original file line number	Diff line number	Diff line change
`@@ -849,7 +849,7 @@ static void vhost_dev_lock_vqs(struct vhost_dev *d)`
`849`	`849`	`{`
`850`	`850`	`int i = 0;`
`851`	`851`	`for (i = 0; i < d->nvqs; ++i)`
`852`		`- mutex_lock(&d->vqs[i]->mutex);`
	`852`	`+ mutex_lock_nested(&d->vqs[i]->mutex, i);`
`853`	`853`	`}`
`854`	`854`
`855`	`855`	`static void vhost_dev_unlock_vqs(struct vhost_dev *d)`