2.6.22-rc1 suspend to RAM problem

2.6.22-rc1 suspend to RAM problem

[Marcus Better]

Hi,

my Thinkpad R60 (x86_64, SMP, XFS) has failed to resume from RAM twice, with
the caps lock LED flashing, blank screen and a locked system (but it
usually succeeds). I use s2ram.

Marcus

Re: 2.6.22-rc1 suspend to RAM problem

[Marcus Better]

Marcus Better wrote:
> my Thinkpad R60 (x86_64, SMP, XFS) has failed to resume from RAM twice,

I should probably add that I have applied the Rafael's patch [1] for the
suspend-to-disk issue.

Marcus

[1]
http://article.gmane.org/gmane.linux.power-management.general/5593/match=wish+machine

Re: Re: 2.6.22-rc1 suspend to RAM problem

[Rafael J. Wysocki]

On Wednesday, 16 May 2007 15:13, Marcus Better wrote:
> Marcus Better wrote:
> > my Thinkpad R60 (x86_64, SMP, XFS) has failed to resume from RAM twice,
> 
> I should probably add that I have applied the Rafael's patch [1] for the
> suspend-to-disk issue.

Please revert it.  It's a debug-only patch and it fiddles with too many things.

Does the hibernation work with it, BTW?

Rafael

Re: Re: 2.6.22-rc1 suspend to RAM problem

[Marcus Better]

[-- Attachment #1.1: Type: text/plain, Size: 234 bytes --]

Rafael J. Wysocki wrote:
> Please revert it.  It's a debug-only patch and it fiddles with too many
> things.
>
> Does the hibernation work with it, BTW?

Yes, that's why I left it in place. Will remove and try again.

Thanks,

Marcus

[-- Attachment #1.2: Type: application/pgp-signature, Size: 189 bytes --]

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

Re: Re: 2.6.22-rc1 suspend to RAM problem

[Marcus Better]

[-- Attachment #1.1: Type: text/plain, Size: 643 bytes --]

> On Wednesday, 16 May 2007 15:13, Marcus Better wrote:
> > Marcus Better wrote:
> > > my Thinkpad R60 (x86_64, SMP, XFS) has failed to resume from RAM twice,
> >
> > I should probably add that I have applied the Rafael's patch [1] for the
> > suspend-to-disk issue.
>
> Please revert it.  It's a debug-only patch and it fiddles with too many
> things.

Ok, so the resume-from-RAM problem appears on 2.6.22-rc1, but not on 2.6.20, 
and probably not in 2.6.21. I may try to bisect it if I can find some time...

Also, it works correctly with "init=/bin/sh", but not when booting 
with "single". Perhaps a driver issue?

Marcus

[-- Attachment #1.2: Type: application/pgp-signature, Size: 189 bytes --]

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

Re: Re: 2.6.22-rc1 suspend to RAM problem

[Pavel Machek]

Hi!

> > > > my Thinkpad R60 (x86_64, SMP, XFS) has failed to resume from RAM twice,
> > >
> > > I should probably add that I have applied the Rafael's patch [1] for the
> > > suspend-to-disk issue.
> >
> > Please revert it.  It's a debug-only patch and it fiddles with too many
> > things.
> 
> Ok, so the resume-from-RAM problem appears on 2.6.22-rc1, but not on 2.6.20, 
> and probably not in 2.6.21. I may try to bisect it if I can find some time...
> 
> Also, it works correctly with "init=/bin/sh", but not when booting 
> with "single". Perhaps a driver issue?

Sounds so. Try to locate which driver.
							Pavel

-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

Re: Re: 2.6.22-rc1 suspend to RAM problem

[Marcus Better]

Pavel Machek wrote:
> Sounds so. Try to locate which driver.

Found it! It's fw-ohci. Haven't bisected it yet though.

Marcus

Re: Re: 2.6.22-rc1 suspend to RAM problem

[Marcus Better]

Marcus Better wrote:
> Found it! It's fw-ohci. Haven't bisected it yet though.

Results so far: Commit 7b104bcb8e460e45a1aebe3da9b86aacdb4cab12 is bad,
b526ca438b95a6d71210e0ffc79aabac8aba2b1e is good, 261 more revisions to
test.

Marcus

Re: Re: 2.6.22-rc1 suspend to RAM problem

[Marcus Better]

Marcus Better wrote:
> Results so far: Commit 7b104bcb8e460e45a1aebe3da9b86aacdb4cab12 is bad,

Forget it, the bisection was borked...

Marcus

Re: Re: 2.6.22-rc1 suspend to RAM problem

[Marcus Better]

Pavel Machek wrote:
> Sounds so. Try to locate which driver.

So it's the fw-ohci driver, but I cannot bisect it since it's based on
2.6.21-rc3 which hangs already at suspend.

Marcus

Re: Re: Re: 2.6.22-rc1 suspend to RAM problem

[Johannes Berg]

[-- Attachment #1.1: Type: text/plain, Size: 382 bytes --]

On Thu, 2007-05-17 at 23:19 +0200, Marcus Better wrote:
> Marcus Better wrote:
> > Results so far: Commit 7b104bcb8e460e45a1aebe3da9b86aacdb4cab12 is bad,
> 
> Forget it, the bisection was borked...

fw-ohci is new anyway and I don't think it has seen many revisions
in-tree. It's loaded on my machine too but lost against ohci1394 so it's
not actually in use.

johannes

[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 190 bytes --]

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

Re: Re: Re: 2.6.22-rc1 suspend to RAM problem

[Pavel Machek]

Hi!

> > Sounds so. Try to locate which driver.
> 
> Found it! It's fw-ohci. Haven't bisected it yet though.

Complain to its maintainer, cc list and akpm, I'd say.
									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html