Re: 2.6.13-rc7-rt1

[Thomas Gleixner <tglx@linutronix.de>]

On Thu, 2005-08-25 at 16:29 -0700, Daniel Walker wrote:
> Devastating latency on a 3Ghz xeon .. Maybe the raw_spinlock in the
> timer base is creating a unbounded latency?

The lock is only held for really short periods. The only possible long
period would be migration of timers from a dead hotplug cpu to another.
I guess thats not the case.

Do you have HIGH_RES_TIMERS enabled ?

> ( softirq-timer/1-13   |#1): new 66088 us maximum-latency critical section.
>  => started at timestamp 1857957769: <__down_mutex+0x5f/0x295>
>  =>   ended at timestamp 1858023857: <_raw_spin_unlock_irq+0x16/0x39>

which mutex was taken and which raw spinlock released  ?

__down_mutex / _raw_spin_unlock_irq is an asymetric pair. Those sections
should be symetric.

I have the feeling that the call trace is not complete.

> <ffffffff8047732a>{rt_secret_rebuild+0}

hint: rt_secret_rebuild() is a well known long running timer callback
function. 

tglx