Re: [patch] voluntary-preempt-2.6.9-rc1-bk4-Q5

[Ingo Molnar]

* Mark_H_Johnson@xxxxxxxxxxxx <Mark_H_Johnson@xxxxxxxxxxxx> wrote:

> >since the latency tracer does not trigger, we need a modified tracer to
> >find out what's happening during such long delays. I've attached the
> >'user-latency-tracer' patch ontop of -Q5, which is a modification of the
> >latency tracer.
> Grr. I should have checked before I built with this patch. With this in
> I now get the
>   kernel: Could not allocate 4 bytes percpu data
> messages again. Need to increase that data area so
>   #define PERCPU_ENOUGH_ROOM 196608
> or something similar (should leave about 50K free for modules).

ok, i've upped that value in my tree too.

> I will rebuild with this change plus the latest of the others.

since your traces do show those 'mystic' latency incidents causing
200-500 usec overhead in few-instructions code paths, perhaps the 1 msec
jitter you are seeing could a consequence of this too. So perhaps it
would make sense to first try disabling IDE DMA, to see whether this has
any effect on the magic latencies.

	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/