Re: [Bug #15096] Resume lock up -- bisected, commit 3a1151e3f124fd1a2c54b8153f510f1a7c715369

[Rafael J. Wysocki]

On Monday 08 February 2010, RafaÅ MiÅecki wrote:
> W dniu 8 lutego 2010 20:24 uÅytkownik Rafael J. Wysocki <rjw@xxxxxxx> napisaÅ:
> > On Monday 08 February 2010, RafaÅ MiÅecki wrote:
> >> W dniu 8 lutego 2010 11:56 uÅytkownik Rafael J. Wysocki <rjw@xxxxxxx> napisaÅ:
> >> > On Monday 08 February 2010, RafaÅ MiÅecki wrote:
> >> >> 2010/2/8 Rafael J. Wysocki <rjw@xxxxxxx>:
> >> >> > This message has been generated automatically as a part of a report
> >> >> > of regressions introduced between 2.6.31 and 2.6.32.
> >> >> >
> >> >> > The following bug entry is on the current list of known regressions
> >> >> > introduced between 2.6.31 and 2.6.32.  Please verify if it still should
> >> >> > be listed and let the tracking team know (either way).
> >> >> >
> >> >> >
> >> >> > Bug-Entry       : http://bugzilla.kernel.org/show_bug.cgi?id=15096
> >> >> > Subject         : Resume lock up -- bisected, commit 3a1151e3f124fd1a2c54b8153f510f1a7c715369
> >> >> > Submitter       : RafaÅ MiÅecki <zajec5@xxxxxxxxx>
> >> >> > Date            : 2010-01-20 23:15 (19 days old)
> >> >> > First-Bad-Commit: http://git.kernel.org/git/linus/3a1151e3f124fd1a2c54b8153f510f1a7c715369
> >> >
> >> > It looks like we should revert this commit, seems broken.
> >> >
> >> > Anyway, I'll do my best to have a look at this bug later today.
> >>
> >> In such case I'd prefer to add "broken_DOS" flag for devices that were
> >> enabled by this patch. We could avoid restoring backlight on resume
> >> for broken_DOS then and maybe just update brightness to current value.
> >
> > Or perhaps we can try to restore the backlight at a different place.
> 
> I wanted to find out what causes this lock up by syncing modules
> resume or putting some delays. See bug report of my question.
> Unfortunately didn't get answer.

Hmm.

I think the JÃrÃme's comment at

http://bugzilla.kernel.org/show_bug.cgi?id=15096#c19

is actually correct.  It looks like acpi_video_resume() is executed before the
KMS driver's resume and things go south.

In fact, we seem to have a design issue with acpi_video_resume(), because it's
called too early.

Rafael
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/