Re: 2.6.14.x and weird things with interrupts on smp machines

From: Arkadiusz Miskiewicz
Date: Sun Jan 08 2006 - 15:20:54 EST


On Sunday 08 January 2006 19:49, Arkadiusz Miskiewicz wrote:
> On Sunday 08 January 2006 19:31, Arkadiusz Miskiewicz wrote:
> > I've recently noticed that something weird is happening on my SMP
> > machines. Both machines are 2 x Xeon CPU with HT enabled.
> > /proc/interrupts shows that only CPU#0 is used which is very weird (and
> > CPU#1 on one of the machines). I'm running userspace irqbalance, too. I'm
> > unable to alter affinity settings for irqs - these are always the same as
> > below.
> >
> > Has anyone noticed such problems?
>
> Seems that not only me:
>
> https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=172909
>
>
> Is this related?
> http://www.nabble.com/smp_affinity-weirdness-in-LK-2.6.14-t496221.html
>
> (since one of users here reports that the problem for him seems to be fixed
> in 2.6.15). If it is then I would love to see it in stable 2.6.14.x
> release.

That patch
http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff_plain;h=fe655d3a06488c8a188461bca493e9f23fc8c448;hp=b0b623c3b22d57d6941b200321779d56c4e79e6b
seems to fix the problem:

# cat /proc/interrupts
CPU0 CPU1 CPU2 CPU3
0: 5580 4005 4004 3100 IO-APIC-edge timer
4: 488 93 3 231 IO-APIC-edge serial
8: 1 0 0 0 IO-APIC-edge rtc
9: 0 0 0 0 IO-APIC-level acpi
169: 2698 0 11062 16885 IO-APIC-level qla2300
177: 63677 0 0 0 IO-APIC-level eth0
185: 6 0 17417 0 IO-APIC-level eth1
NMI: 16778 16762 16758 16760
LOC: 16515 16561 16680 16704
ERR: 0
MIS: 0

Please put in in stable 2.6.14.x since it's quite important bugfix.

--
Arkadiusz Miśkiewicz PLD/Linux Team
http://www.t17.ds.pwr.wroc.pl/~misiek/ http://ftp.pld-linux.org/
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/