ACPI Suspend, devd and rc.suspend

[Norberto Meijome]

On Mon, 27 Mar 2006 08:33:26 -0800
"Kevin Oberman" <oberman at es.net> wrote:

> > Date: Mon, 27 Mar 2006 23:35:09 +1000
> > From: Norberto Meijome <freebsd at meijome.net>
> > Sender: owner-freebsd-mobile at freebsd.org
> > 
> > Hi there,
> > machine : Thinkpad z60m, acpi.ko, acpi_ibm.ko loaded, 6.1 Prerelease
> > kernel and world as of yesterday.
> > 
> > If I run zzz , /etc/rc.suspend is run. on resume, /etc/rc.resume is
> > executed.
> > 
> > If I hit Fn-F4 ( == suspend event), the acpi sets the machine to
> > suspend mode, but /etc/rc.suspend is NOT executed, and neither
> > is /etc/rc.resume on resume.
> > 
> > I don't know how to tell if devd catches the suspend event. I run
> > it in debug mode (devd -dD) but couldn't see anything other than
> > all the device motherboard's USB hubs being pulled off and added
> > back in.
> > 
> > I would love some enlightenment on this subject.
> 
> In HEAD, devd should be firing off rc.resume, but the problem is in
> the suspend side. When triggered by Fn-F4, there is no way to be sure
> that rc.suspend is complete before the system halts. zzz (or acpiconf
> -s3) will complete rc.suspend before actually telling the system to
> suspend. On the other hand, Fn-F4 immediately starts the suspend
> sequence. 
> 
> While I believe that it can be set to wait for a specific period
> before suspending, I don't think that there is any way to assure that
> anything completes.
> 
> njl@ or Bruno Ducrot would know more about the details. This has been
> discussed in the past; probably in acpi at .

Kevin, thanks again for the explanation and time to reply. I figured
the BIOS was taking over at some point.

Fabian Keil's reply to my other email to the list (thread before this
one) points me into a possible solution.

thx
beto