Hi.

On Sun, 2007-05-27 at 23:45 +0200, Rafael J. Wysocki wrote:
quoted text
> On Sunday, 27 May 2007 22:49, Matthew Garrett wrote:
> > On Sun, May 27, 2007 at 10:31:53PM +0200, Rafael J. Wysocki wrote:
> > > From: Rafael J. Wysocki <rjw@sisk.pl>
> > >=20
> > > Use a hibernation and suspend notifier to disable the firmware reques=
ting
quoted text
> > > mechanism before a hibernation/suspend and enable it after the operat=
ion.
quoted text
> >=20
> > This avoids the problem of .resume methods calling userspace while=20
> > userspace is frozen and a resulting hang, but does it actually result i=
n=20
quoted text
> > the drivers beginning to work again?
>=20
> Well, this was acutally invented before you've decided to remove the free=
zing
quoted text
> of tasks from the suspend code path (which I think is a mistake, but that=
's
quoted text
> only my personal opinion, so it doesn't matter very much ;-)) and I regar=
d it
quoted text
> as a workaround.

Suspend-to-ram code paths shouldn't assume userspace is unfrozen anyway.
Doesn't [u]swsusp have a code path like Suspend2 where we can suspend to
ram after writing the hibernation image? In that case, it will still be
possible that we seek to enter and leave S3 with processes frozen.

Apologies if anyone has already mentioned this - I'm just starting to
play catchup.

Regards,

Nigel