Re: [BISECTED] 2.6.35.*: horrible (exponential? >linear) slowdown to unusability (HPET)

[Artur Skawina <art.08.09@gmail.com>]

On 09/10/10 11:41, Jiri Slaby wrote:
> AFAIU, clocksource=jiffies shouldn't fix this issue. If that's the case
> we have another 2.6.32->2.6.33 regression (and I will report it
> separately). People report the system is unusable (sleepers are not
> woken), unless clocksource=jiffies, clocksource=tsc or nolapic_timer is
> used or a key pressed (i.e. some HW interrupt):
> https://bugzilla.novell.com/show_bug.cgi?id=579932
> (there is also a report with 2.6.35.3 vanilla)

I saw this first in v2.6.35-rc4. The symptom is that suddenly everything
stops and the only way to make things progress is to generate some events.
Apps that are waiting for a timeout aren't woken up unless you press a key
etc. After a few minutes (and lots of key presses ;) ) the machine usually
recovers, leaving no trace in logs.
It's a nasty bug to debug as there's no way to trigger it, at least here,
and it usually does not want to happen when you most want it to; took me
~week to strike a bios upgrade from the suspect list, another one to
exclude intel_idle and so on. Never got around to check 30a564be9d955,
which was on the list too, unfortunately. I'll bet it /is/ the cause, but
won't be able to confirm this for quite a while...

artur