Re: Intermittent early panic in try_to_wake_up

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Sunday 08 November 2009, Pavel Machek wrote:
> 
> > > > if I copied the entire panic message entirely, including the Code section?
> > > > I can try that the next time it happens.
> > > 
> > > Use vga=ask boot_delay=100 select the highest res possible.
> > > 
> > > Possibly you could use a digital (video) camera to record the output.
> > 
> > FWIW, I see a crash in autoremove_wake_function() during resume from suspend
> > to RAM that happens immediately before the thawing of tasks.  It always happens
> > in the "events" thread.  I can trigger it quite reliably, although not 100% of
> > the time.
> > 
> > I can only see the bottom half of the call trace, so I can't really say what
> > kind of a crash this is.
> 
> Our oops printing code sucks. It should mdelay after each screen or
> something like that. Call trace should be in reverse order, so the
> most useful parts stay on the screen...

In this particular case it should fit into the screen, but it happens during
a VT switch or something like this.

Unfortunately, this issue turns out to be quite difficult to nail down and only
triggers in specific configurations.

Thanks,
Rafael