lkml.org 
[lkml]   [2008]   [Jul]   [21]   [last100]   RSS Feed
Views: [more markup]  [less markup]  [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
DateMon, 21 Jul 2008 08:35:27 -0700 (PDT)
SubjectRe: [GIT]: Networking
FromDavid Miller <>
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Mon, 21 Jul 2008 11:16:27 +0100

> Alexander Beregalov wrote:
> > Hi David!
> > 
> > There is another problem on sparc64 and happymeal ethernet card.
> > 
> > when tring to up interface:
> > 
> > kernel BUG at net/core/dev.c:1328
> [...]
> 
> This is yet another driver calling netif_wake_queue() during dev_open(),
> when there is no real qdisc present.  (And yes, sfc is another of those
> drivers - I will post a patch after internal review.)

Yep, what idiot wrote this driver? ;-)

Alexander please try this patch:

sunhme: Remove stop/wake TX queue calls in set-multicast-list handler.

Based upon a bug report by Alexander Beregalov and commentary
from Ben Hutchings.

These are totally unnecessary, in particular because this
driver's ->hard_start_xmit() handler takes the same driver
spinlock that the set-multicast-list handler uses.

Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/net/sunhme.c b/drivers/net/sunhme.c
index 1aa425b..b79d5f0 100644
--- a/drivers/net/sunhme.c
+++ b/drivers/net/sunhme.c
@@ -2377,8 +2377,6 @@ static void happy_meal_set_multicast(struct net_device *dev)
 
 	spin_lock_irq(&hp->happy_lock);
 
-	netif_stop_queue(dev);
-
 	if ((dev->flags & IFF_ALLMULTI) || (dev->mc_count > 64)) {
 		hme_write32(hp, bregs + BMAC_HTABLE0, 0xffff);
 		hme_write32(hp, bregs + BMAC_HTABLE1, 0xffff);
@@ -2410,8 +2408,6 @@ static void happy_meal_set_multicast(struct net_device *dev)
 		hme_write32(hp, bregs + BMAC_HTABLE3, hash_table[3]);
 	}
 
-	netif_wake_queue(dev);
-
 	spin_unlock_irq(&hp->happy_lock);
 }
 

\
 
 \ /
  Last update: 2008-07-21 17:39    [from the cache]
©2003-2008