Skip to contentSkip to conversation

bnx2: Use netif_carrier_off() to prevent timeout.

PATCH Applied Michael Chan 2 participants 2 messages May 13, 2010 – May 18, 2010

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 bnx2: Use netif_carrier_off() to prevent timeout.

Based on original patch from Stanislaw Gruszka <[email protected]>.

Using netif_carrier_off() is better than updating all the ->trans_start on all the tx queues.

netif_carrier_off() needs to be called after bnx2_disable_int_sync() to guarantee no race conditions with the serdes timers that can modify the carrier state.

If the chip or phy is reset, carrier will turn back on when we get the link interrupt. If there is no reset, we need to turn carrier back on in bnx2_netif_start(). Again, the phy_lock prevents race conditions with the serdes timers.

---
 drivers/net/bnx2.c |   14 +++++---------
 1 files changed, 5 insertions(+), 9 deletions(-)
drivers/net/bnx2.c +5 -9
index ab0068c..188e356 100644
@@ -656,19 +656,11 @@ bnx2_netif_stop(struct bnx2 *bp, bool stop_cnic)
 	if (stop_cnic)
 		bnx2_cnic_stop(bp);
 	if (netif_running(bp->dev)) {
-		int i;
-
 		bnx2_napi_disable(bp);
 		netif_tx_disable(bp->dev);
-		/* prevent tx timeout */
-		for (i = 0; i <  bp->dev->num_tx_queues; i++) {
-			struct netdev_queue *txq;
-
-			txq = netdev_get_tx_queue(bp->dev, i);
-			txq->trans_start = jiffies;
-		}
 	}
 	bnx2_disable_int_sync(bp);
+	netif_carrier_off(bp->dev);	/* prevent tx timeout */
 }
 
 static void
@@ -677,6 +669,10 @@ bnx2_netif_start(struct bnx2 *bp, bool start_cnic)
 	if (atomic_dec_and_test(&bp->intr_sem)) {
 		if (netif_running(bp->dev)) {
 			netif_tx_wake_all_queues(bp->dev);
+			spin_lock_bh(&bp->phy_lock);
+			if (bp->link_up)
+				netif_carrier_on(bp->dev);
+			spin_unlock_bh(&bp->phy_lock);
 			bnx2_napi_enable(bp);
 			bnx2_enable_int(bp);
 			if (start_cnic)
Signature · 1 line
1.6.4.GIT
maintainer Applied
#1 lore raw

From: "Michael Chan" <[email protected]>
Date: Thu, 13 May 2010 07:48:35 -0700

Quoted 9 lines

Based on original patch from Stanislaw Gruszka <[email protected]>.

Using netif_carrier_off() is better than updating all the ->trans_start on all the tx queues.

netif_carrier_off() needs to be called after bnx2_disable_int_sync() to guarantee no race conditions with the serdes timers that can modify the carrier state.

Quoted 7 lines

If the chip or phy is reset, carrier will turn back on when we get the link interrupt. If there is no reset, we need to turn carrier back on in bnx2_netif_start(). Again, the phy_lock prevents race conditions with the serdes timers.

Applied, thanks.