[Fwd: Re: Plea for help]

[Fwd: Re: Plea for help]

[Ian Pilcher]

Yu, Luming wrote:
> 
> Since I cannot reporduce it on my side. Would you please dig into 
> function acpi_ev_create_gpe_block.
> 

Before I sent the first message in this thread, I added a bunch of
printks to this function.  They ended up mixed in with the panic, just
like the ACPI debugging messages.

I'm not sure what else I can do to "dig into" it.

> 
>>And just for grins and giggles, I tried booting with 'idle=poll'.  Sure
>>enough, no panic.
> 
> 
> I'm confused! 
> 

My guess, and that's all it can be, is that the transition to ACPI mode
does something that causes the idle thread on the other processor to
crash.  Perhaps a spurious interrupt is being thrown, or an interrupt is
routed to the wrong processor.  (Is the behavior of a HLTed processor
that gets a misrouted interrupt different than that of a processor that
is polling?)

There certainly seems to be a timing sensitivity too.  Is there a kernel
mode sleep-like function that I could play with?

Thanks!

-- 
========================================================================
Ian Pilcher                                        i.pilcher-Wuw85uim5zDR7s880joybQ@public.gmane.org
========================================================================



-------------------------------------------------------
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click

RE: [Fwd: Re: Plea for help]

[Yu, Luming]

>> 
>> Since I cannot reporduce it on my side. Would you please dig into 
>> function acpi_ev_create_gpe_block.
>> 
>Before I sent the first message in this thread, I added a bunch of
>printks to this function.  They ended up mixed in with the panic, just
>like the ACPI debugging messages.
>
>I'm not sure what else I can do to "dig into" it.

Maybe you need to file a tracker on bugzilla.kernel.org.
You can put anything you found to let us enjoy.
It's an effective  way.

> Perhaps a spurious interrupt is being thrown, or an interrupt is
>routed to the wrong processor.  (Is the behavior of a HLTed processor
>that gets a misrouted interrupt different than that of a processor that
>is polling?)

It's a hint.

>There certainly seems to be a timing sensitivity too.  Is there a
kernel
>mode sleep-like function that I could play with?

wait_event?


-------------------------------------------------------
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id\x1278&alloc_id371&op=click

Re: [Fwd: Re: Plea for help]

[Ian Pilcher]

Yu, Luming wrote:
> 
> 
>>There certainly seems to be a timing sensitivity too.  Is there a
> 
> kernel
> 
>>mode sleep-like function that I could play with?
> 
> 
> wait_event?
> 

Can you point me to an example of using wait_event to "sleep" for a set
period of time?

Thanks!

-- 
========================================================================
Ian Pilcher                                        i.pilcher-Wuw85uim5zDR7s880joybQ@public.gmane.org
========================================================================



-------------------------------------------------------
This SF.net email is sponsored by: Perforce Software.
Perforce is the Fast Software Configuration Management System offering
advanced branching capabilities and atomic changes on 50+ platforms.
Free Eval! http://www.perforce.com/perforce/loadprog.html

RE: [Fwd: Re: Plea for help]

[Yu, Luming]

> 
> Can you point me to an example of using wait_event to "sleep" 
> for a set
> period of time?
> 

You need to init a timer for timing. Wait_event just let you sleep.
When timeout, function wake_up could be used to complete sleeping.


-------------------------------------------------------
This SF.net email is sponsored by: Perforce Software.
Perforce is the Fast Software Configuration Management System offering
advanced branching capabilities and atomic changes on 50+ platforms.
Free Eval! http://www.perforce.com/perforce/loadprog.html