Re: [Bug #12465] KVM guests stalling on 2.6.28 (bisected)

[Ingo Molnar]

* Kevin Shanahan <kmshanah@xxxxxxxxxxx> wrote:

> Running the ping test with without apache2 running in the guest:
> 
> --- hermes-old.wumi.org.au ping statistics ---
> 900 packets transmitted, 900 received, 0% packet loss, time 902740ms
> rtt min/avg/max/mdev = 0.568/3.745/272.558/16.990 ms
> 
> And with apache2 running:
> 
> --- hermes-old.wumi.org.au ping statistics ---
> 900 packets transmitted, 900 received, 0% packet loss, time 902758ms
> rtt min/avg/max/mdev = 0.625/25.634/852.739/76.586 ms
> 
> In both cases it's quite variable, but the max latency is still not as 
> bad as when running with the irq chip enabled.

So the worst-case ping latency is more than 10 times lower?

I'd say this points in the direction of some sort of KVM-internal 
wakeup/signalling latency that happens if KVM does not deschedule. For 
example it could be a bug like this: if a guest image runs at 100% CPU 
time for a long time, IRQ injections might not propagate up until the 
preemption callbacks run. (but i'm just speculating here)

	Ingo
--
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/