IPoIB: Make ipoib_mcast_stop_thread flush the workqueue

We used to pass a flush variable to mcast_stop_thread to indicate if
we should flush the workqueue or not.  This was due to some code
trying to flush a workqueue that it was currently running on which is
a no-no.  Now that we have per-device work queues, and now that
ipoib_mcast_restart_task has taken the fact that it is queued on a
single thread workqueue with all of the ipoib_mcast_join_task's and
therefore has no need to stop the join task while it runs, we can do
away with the flush parameter and unilaterally flush always.

Signed-off-by: Doug Ledford <[email protected]>
Signed-off-by: Roland Dreier <[email protected]>

drivers/infiniband/ulp/ipoib/ipoib.h

@@ -499,7 +499,7 @@ void ipoib_mcast_send(struct net_device *dev, u8 *daddr, struct sk_buff *skb);
 
 void ipoib_mcast_restart_task(struct work_struct *work);
 int ipoib_mcast_start_thread(struct net_device *dev);
-int ipoib_mcast_stop_thread(struct net_device *dev, int flush);
+int ipoib_mcast_stop_thread(struct net_device *dev);
 
 void ipoib_mcast_dev_down(struct net_device *dev);
 void ipoib_mcast_dev_flush(struct net_device *dev);

drivers/infiniband/ulp/ipoib/ipoib_ib.c

@@ -747,7 +747,7 @@ int ipoib_ib_dev_down(struct net_device *dev, int flush)
 	clear_bit(IPOIB_FLAG_OPER_UP, &priv->flags);
 	netif_carrier_off(dev);
 
-	ipoib_mcast_stop_thread(dev, flush);
+	ipoib_mcast_stop_thread(dev);
 	ipoib_mcast_dev_flush(dev);
 
 	ipoib_flush_paths(dev);
@@ -1097,7 +1097,7 @@ void ipoib_ib_dev_cleanup(struct net_device *dev)
 	 */
 	ipoib_flush_paths(dev);
 
-	ipoib_mcast_stop_thread(dev, 1);
+	ipoib_mcast_stop_thread(dev);
 	ipoib_mcast_dev_flush(dev);
 
 	ipoib_transport_dev_cleanup(dev);

drivers/infiniband/ulp/ipoib/ipoib_multicast.c

@@ -648,7 +648,7 @@ int ipoib_mcast_start_thread(struct net_device *dev)
 	return 0;
 }
 
-int ipoib_mcast_stop_thread(struct net_device *dev, int flush)
+int ipoib_mcast_stop_thread(struct net_device *dev)
 {
 	struct ipoib_dev_priv *priv = netdev_priv(dev);
 
@@ -659,8 +659,7 @@ int ipoib_mcast_stop_thread(struct net_device *dev, int flush)
 	cancel_delayed_work(&priv->mcast_task);
 	mutex_unlock(&mcast_mutex);
 
-	if (flush)
-		flush_workqueue(priv->wq);
+	flush_workqueue(priv->wq);
 
 	return 0;
 }
@@ -838,8 +837,6 @@ void ipoib_mcast_restart_task(struct work_struct *work)
 
 	ipoib_dbg_mcast(priv, "restarting multicast task\n");
 
-	ipoib_mcast_stop_thread(dev, 0);
-
 	local_irq_save(flags);
 	netif_addr_lock(dev);
 	spin_lock(&priv->lock);
@@ -936,13 +933,10 @@ void ipoib_mcast_restart_task(struct work_struct *work)
 	 * We have to cancel outside of the spinlock, but we have to
 	 * take the rtnl lock or else we race with the removal of
 	 * entries from the remove list in mcast_dev_flush as part
-	 * of ipoib_stop() which will call mcast_stop_thread with
-	 * flush == 1 while holding the rtnl lock, and the
-	 * flush_workqueue won't complete until this restart_mcast_task
-	 * completes.  So do like the carrier on task and attempt to
-	 * take the rtnl lock, but if we can't before the ADMIN_UP flag
-	 * goes away, then just return and know that the remove list will
-	 * get flushed later by mcast_stop_thread.
+	 * of ipoib_stop().  We detect the drop of the ADMIN_UP flag
+	 * to signal that we have hit this particular race, and we
+	 * return since we know we don't need to do anything else
+	 * anyway.
 	 */
 	while (!rtnl_trylock()) {
 		if (!test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags))
@@ -954,6 +948,9 @@ void ipoib_mcast_restart_task(struct work_struct *work)
 		ipoib_mcast_leave(mcast->dev, mcast);
 		ipoib_mcast_free(mcast);
 	}
+	/*
+	 * Restart our join task if needed
+	 */
 	ipoib_mcast_start_thread(dev);
 	rtnl_unlock();
 }