On Tue, Mar 22, 2005 at 12:10:14AM +0100, Magnus Naeslund(t) wrote:
> Ingo Molnar wrote:
> >
> >i've uploaded my current tree (-V0.7.41-01) to:
> >
> >  http://redhat.com/~mingo/realtime-preempt/
> >
> >it includes the latest round of RCU fixes - but doesnt solve the SMP
> >bootup crash.
> >
> >     Ingo
> 
> With this kernel I can run for some 20 minutes, then the ip_dst_cache 
> overflows.
> I gather it has something to do with RCU...
> 
> If I just let it run and grep ip_dst_cache /proc/slab it goes up to 4096 
> (max) and then the printk's are starting, and the networks stops.
> After i up the limit to the double (echo "8192" > 
> /proc/sys/net/ipv4/route/max_size) network starts to work again.
> But of course, after a while it overflows again:
> 
> # grep ip_dst_cache /proc/slabinfo
> ip_dst_cache        8192   8205    256   15    1 : tunables   16    8 
>  0 : slabdata    547    547      0
> 
> I haven't tried the vanilla 2.6.12-rc2 kernel, and I don't have the time 
> to do that right now, but i figured it would have something to do with 
> your patch. Older 2.6.8 works just fine.

Hello, Magnus,

I believe that my earlier patch might take care of this (included again
for convenience).

                                                Thanx, Paul

Signed-off-by: <[EMAIL PROTECTED]>

diff -urpN -X dontdiff linux-2.6.11.fixes/kernel/rcupdate.c 
linux-2.6.11.fixes2/kernel/rcupdate.c
--- linux-2.6.11.fixes/kernel/rcupdate.c        Mon Mar 21 08:14:47 2005
+++ linux-2.6.11.fixes2/kernel/rcupdate.c       Mon Mar 21 08:17:00 2005
@@ -620,7 +620,7 @@ static void rcu_process_callbacks(void)
                return;
        }
        rdp->donelist = NULL;
-       rdp->donetail = &rdp->waitlist;
+       rdp->donetail = &rdp->donelist;
        put_cpu_var(rcu_data);
        while (list) {
                next = list->next;
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Reply via email to