Realtime Preemption - 2.6.12-final-RT-V0.7.50-35

Realtime Preemption - 2.6.12-final-RT-V0.7.50-35

[Chuck Harding]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 676 bytes --]

still having sox hang with no messages about what is going on. I
have CONFIG_DEBUG_PREEMPT enabled. It did work without hanging for
about 50 times of playing a sound file, so the problem seems to take
a bit of time to develop. What other info would you need to figure
this out?

-- 
Charles D. (Chuck) Harding <charding@llnl.gov>  Voice: 925-423-8879
Senior Computer Associate         ICCD            Fax: 925-423-6961
Lawrence Livermore National Laboratory      Computation Directorate
Livermore, CA USA  http://www.llnl.gov  GPG Public Key ID: B9EB6601
------------------ http://tinyurl.com/5w5ey -----------------------
-- Fer sell cheep: IBM spel chekker. Wurks grate. --

[-- Attachment #2: 2.6.12-RT-V0.7.50-35 config --]
[-- Type: APPLICATION/x-gzip, Size: 8264 bytes --]

Re: Realtime Preemption - 2.6.12-final-RT-V0.7.50-35

[Ingo Molnar]

* Chuck Harding <charding@llnl.gov> wrote:

> still having sox hang with no messages about what is going on. I have 
> CONFIG_DEBUG_PREEMPT enabled. It did work without hanging for about 50 
> times of playing a sound file, so the problem seems to take a bit of 
> time to develop. What other info would you need to figure this out?

could you chrt the sound IRQ thread to SCHED_OTHER (audio performance 
will suck, but it will be more debuggable) - when the lockup happens, do 
you see the IRQ thread looping? Do you have SOFTLOCKUP_DETECT turned on 
too?

	Ingo

Re: Realtime Preemption - 2.6.12-final-RT-V0.7.50-35

[Chuck Harding]

On Wed, 29 Jun 2005, Ingo Molnar wrote:

>
> * Chuck Harding <charding@llnl.gov> wrote:
>
>> still having sox hang with no messages about what is going on. I have
>> CONFIG_DEBUG_PREEMPT enabled. It did work without hanging for about 50
>> times of playing a sound file, so the problem seems to take a bit of
>> time to develop. What other info would you need to figure this out?
>
> could you chrt the sound IRQ thread to SCHED_OTHER (audio performance
> will suck, but it will be more debuggable) - when the lockup happens, do
> you see the IRQ thread looping? Do you have SOFTLOCKUP_DETECT turned on
> too?
>
> 	Ingo

Did that, the IRQ thread is not looping, it doesn't seem to be using any
CPU at all. And I do have SOFTLOCKUP_DETECT on.

(PS, I am subbed to the list so please, no need to CC me. Thanks)

-- 
Charles D. (Chuck) Harding <charding@llnl.gov>  Voice: 925-423-8879
Senior Computer Associate         ICCD            Fax: 925-423-6961
Lawrence Livermore National Laboratory      Computation Directorate
Livermore, CA USA  http://www.llnl.gov  GPG Public Key ID: B9EB6601
------------------ http://tinyurl.com/5w5ey -----------------------
-- Oxymoron: Weather Forecast. --