Re: [gpm] FC5 troubles: how do acpi, pm-utils, and gnome-power-manager fit together?

[Richard Hughes <hughsient gmail com>]

On Tue, 2006-05-16 at 12:51 -0500, Paul Johnson wrote:
> On 5/16/06, Richard Hughes <hughsient gmail com> wrote:
> > And add the blacklisted modules to SUSPEND_MODULES
> 
> Can you give me a hint about the syntax. Currently, /etc/pm/config
> only has these two lines,
> 
> $ cat /etc/pm/config
> SUSPEND_MODULES="button"
> 
> HIBERNATE_RESUME_POST_VIDEO="no"
> 
> and I cannot find any documentation on the format for entering new
> information here--comma separated, etc.

I think specified by space, we should probably add this as a comment in
the file. Pls tell me if a space works and I'll commit a fix.

> >
> > > i can try to configure that in /etc/acpi/actions/sleep.sh and then the
> > > FN-suspend key combination uses the ACPI script to suspend.  However,
> > > I can't figure how to make gnome-power-manager use that script. GPM
> > > seems always to want pm-suspend.  Yes?
> >
> > Look in hal-system-power-suspend to see the logic in what script gets
> > used.
> >
> Aha. It is all starting to come clear to me.  For the past several
> months, I've wandered far and wide in the internet, taking advice from
> everybody who claimed suspend did work.  That means I've experimented
> with "hibernate", "acpitools", "swsusp2" kernel patches, the ACPI
> scripts I mentioned before, and who knows what else.  Now I see I was
> chasing fragments of various different approaches, none of which were
> entirely consistent with your idea of a coherent pm-utils strategy.

Thanks, I think it all needs to be cleared up one way or another.

> > > On the D800, pm-hibernate worked without trouble, unlike suspend to
> > > RAM, which was fraught with trouble (tough to remove all bad modules
> > > and wake up video dependably).
> > > On this system, if I run pm-hibernate, the system tries to turn off,
> > > it writes on the disk a while, and then the screen goes dark but the
> > > power does not turn off.  The power light stays on, but nothing
> > > happens.  So there's something at the very last minute that fails in
> > > the hibernation.
> >
> > Smeels like a kernel problem to me. No error messages?
> >
> No, there is nothing in /var/log/acpid or /var/log/messages.  Just a
> transition from the shutdown to a restart.  I tried hibernate at
> 11:12, and watched the system not-shutoff until 11:15 to hit the power
> button and force a re-start.
> 
> May 15 11:12:30 pols125 gnome-power-manager: Hibernating computer
> because user clicked hibernate from tray menu
> May 15 11:12:33 pols125 sdpd[1851]: terminating...
> May 15 11:12:33 pols125 hcid[1848]: Exit.
> May 15 11:12:35 pols125 kernel: Freezing cpus ...
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 0
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 1
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 3
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 4
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 5
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 7
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 9
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 11
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 12
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 13
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 16
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 17
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 19
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 20
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 21
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 22
> May 15 11:12:35 pols125 kernel: Breaking affinity for irq 23
> May 15 11:12:35 pols125 kernel: CPU 1 is now offline
> May 15 11:15:00 pols125 syslogd 1.4.1: restart.
> May 15 11:15:00 pols125 kernel: klogd 1.4.1, log source = /proc/kmsg started.
> May 15 11:15:00 pols125 kernel: Linux version 2.6.16-1.2111_FC5smp
> (bhcompile hs20-bc1-3 build redhat com) (gcc version 4.1.0 20060304
> (Red Hat 4.1.0-3)) #1 SMP Thu May 4 21:35:09 EDT 2006
> May 15 11:15:00 pols125 kernel: BIOS-provided physical RAM map:

Hmm. Maybe try lkml? Doesn't look like anything I know about...

> > > So maybe I should work on customizing some ACPI script for that?
> > >
> > > ps.  While testing, I've added the kernel option agp=off and I've
> > > turned on the Nvidia AGP support in xorg.conf.  That was vital on the
> > > D800 and I just did it here after suspend troubles started to arise.
> >
> > There's a pm-utils mailing list[1] where we are discussing the next
> > generation pm-utils replacement.
> >
> > Note, there is a presentation I'm working on about how pm-utils, g-p-m
> > and hal fit together, available on my website, hughsie.com -- that might
> > explain things better.
> 
> That is very nicely done.  For the first time, I understand the appeal
> of HAL.

Cool! I'll add more to the document as stuff changes.

>  Don't you worry that somebody who works on HAL will make a
> mistake or insert some mischief and then nothing at all will work
> anymore?  Without HAL, at least some stuff worked ...

Yes, but the same could be said about bios upgrades, kernel updates,
glibc updates etc.. :-)

Richard.