[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20080818091011.GC5434@ff.dom.local>
Date: Mon, 18 Aug 2008 09:10:12 +0000
From: Jarek Poplawski <jarkao2@...il.com>
To: David Miller <davem@...emloft.net>
Cc: netdev@...r.kernel.org
Subject: Re: [PATCH]: Schedule correct qdisc in watchdog.
On Mon, Aug 18, 2008 at 01:39:54AM -0700, David Miller wrote:
>
> Jarek, I bet this is what was causing the crash you were trying to fix
> yesterday.
>
> But that's just my newbie hunch :-)
Yes, this was one of the reasons, but not the only one, and my oldbie
version of this fix could be found here as "02-fix1.patch":
http://permalink.gmane.org/gmane.linux.network/103039
It's a bit different, because I think wd->qdisc can probably
point to the noop_qdisc (if it's the root qdisc).
Anyway this kind of change is necessary here, and everywhere where
netif_schedule() could be called (I've found sch_cbq.c only).
Jarek P.
>
> pkt_sched: Never scheduler non-root qdiscs.
>
> The qdisc watchdogs can be attached to any qdisc,
> not just the root, so make sure we schedule the
> correct one.
>
> Signed-off-by: David S. Miller <davem@...emloft.net>
>
> diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
> index c8dc72e..98c0084 100644
> --- a/net/sched/sch_api.c
> +++ b/net/sched/sch_api.c
> @@ -426,7 +426,7 @@ static enum hrtimer_restart qdisc_watchdog(struct hrtimer *timer)
>
> wd->qdisc->flags &= ~TCQ_F_THROTTLED;
> smp_wmb();
> - __netif_schedule(wd->qdisc);
> + __netif_schedule(qdisc_root(wd->qdisc));
>
> return HRTIMER_NORESTART;
> }
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists