IPoIB: Refresh paths instead of flushing them on SM change events

The patch tries to solve the problem of device going down and paths being
flushed on an SM change event. The method is to mark the paths as candidates for
refresh (by setting the new valid flag to 0), and wait for an ARP
probe a new path record query.

The solution requires a different and less intrusive handling of SM
change event. For that, the second argument of the flush function
changes its meaning from a boolean flag to a level.  In most cases, SM
failover doesn't cause LID change so traffic won't stop.  In the rare
cases of LID change, the remote host (the one that hadn't changed its
LID) will lose connectivity until paths are refreshed. This is no
worse than the current state.  In fact, preventing the device from
going down saves packets that otherwise would be lost.

Signed-off-by: Moni Levy <monil@voltaire.com>
Signed-off-by: Moni Shoua <monis@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>

1 files changed, 27 insertions, 15 deletions

diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
index 5d50e5261ee..66cafa20c24 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
@@ -902,7 +902,8 @@ int ipoib_ib_dev_init(struct net_device *dev, struct ib_device *ca, int port)
 	return 0;
 }
 
-static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, int pkey_event)
+static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv,
+				enum ipoib_flush_level level)
 {
 	struct ipoib_dev_priv *cpriv;
 	struct net_device *dev = priv->dev;
@@ -915,7 +916,7 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, int pkey_event)
 	 * the parent is down.
 	 */
 	list_for_each_entry(cpriv, &priv->child_intfs, list)
-		__ipoib_ib_dev_flush(cpriv, pkey_event);
+		__ipoib_ib_dev_flush(cpriv, level);
 
 	mutex_unlock(&priv->vlan_mutex);
 
@@ -929,7 +930,7 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, int pkey_event)
 		return;
 	}
 
-	if (pkey_event) {
+	if (level == IPOIB_FLUSH_HEAVY) {
 		if (ib_find_pkey(priv->ca, priv->port, priv->pkey, &new_index)) {
 			clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags);
 			ipoib_ib_dev_down(dev, 0);
@@ -947,11 +948,15 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, int pkey_event)
 		priv->pkey_index = new_index;
 	}
 
-	ipoib_dbg(priv, "flushing\n");
+	if (level == IPOIB_FLUSH_LIGHT) {
+		ipoib_mark_paths_invalid(dev);
+		ipoib_mcast_dev_flush(dev);
+	}
 
-	ipoib_ib_dev_down(dev, 0);
+	if (level >= IPOIB_FLUSH_NORMAL)
+		ipoib_ib_dev_down(dev, 0);
 
-	if (pkey_event) {
+	if (level == IPOIB_FLUSH_HEAVY) {
 		ipoib_ib_dev_stop(dev, 0);
 		ipoib_ib_dev_open(dev);
 	}
@@ -961,27 +966,34 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, int pkey_event)
 	 * we get here, don't bring it back up if it's not configured up
 	 */
 	if (test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags)) {
-		ipoib_ib_dev_up(dev);
+		if (level >= IPOIB_FLUSH_NORMAL)
+			ipoib_ib_dev_up(dev);
 		ipoib_mcast_restart_task(&priv->restart_task);
 	}
 }
 
-void ipoib_ib_dev_flush(struct work_struct *work)
+void ipoib_ib_dev_flush_light(struct work_struct *work)
+{
+	struct ipoib_dev_priv *priv =
+		container_of(work, struct ipoib_dev_priv, flush_light);
+
+	__ipoib_ib_dev_flush(priv, IPOIB_FLUSH_LIGHT);
+}
+
+void ipoib_ib_dev_flush_normal(struct work_struct *work)
 {
 	struct ipoib_dev_priv *priv =
-		container_of(work, struct ipoib_dev_priv, flush_task);
+		container_of(work, struct ipoib_dev_priv, flush_normal);
 
-	ipoib_dbg(priv, "Flushing %s\n", priv->dev->name);
-	__ipoib_ib_dev_flush(priv, 0);
+	__ipoib_ib_dev_flush(priv, IPOIB_FLUSH_NORMAL);
 }
 
-void ipoib_pkey_event(struct work_struct *work)
+void ipoib_ib_dev_flush_heavy(struct work_struct *work)
 {
 	struct ipoib_dev_priv *priv =
-		container_of(work, struct ipoib_dev_priv, pkey_event_task);
+		container_of(work, struct ipoib_dev_priv, flush_heavy);
 
-	ipoib_dbg(priv, "Flushing %s and restarting its QP\n", priv->dev->name);
-	__ipoib_ib_dev_flush(priv, 1);
+	__ipoib_ib_dev_flush(priv, IPOIB_FLUSH_HEAVY);
 }
 
 void ipoib_ib_dev_cleanup(struct net_device *dev)