[ACPI] linux 2.5.67, no suspend state

[B. <Eri...@lp...>]

Hello,

I have a shuttle xpc sb51G (intel chipset) with redhat 9 + kernel 2.5.67
with a patch to drivers/net/8139too.c to help S3 working (the patch was
found on this list. It consists in one call to refrigerator and two calls
to pci_set_p=F4wer_state.)

When I try to go S1, it looks like the computer goes into sleep state and
wakes up immediately. Here is an extract of the logs:

[...]
=3Dzsh entered refrigerator
=3Dsu entered refrigerator
=3Danacron entered refrigerator
=3D|
Suspending devices
Suspending device c03730cc
Suspending devices
Suspending device c03730cc
suspending: hda <0>Suspending devices
uhci-hcd 00:1d.0: suspend to state 1
uhci-hcd 00:1d.1: suspend to state 1
uhci-hcd 00:1d.2: suspend to state 1
ehci-hcd 00:1d.7: suspend to state 1
Suspending device c03730cc
 hwsleep-0257 [18] acpi_enter_sleep_state: Entering sleep state [S1]
Back to C!
Devices Resumed
   evgpe-0396: *** Error: acpi_ev_gpe_dispatch: No handler or method for
GPE[ 0], disabling event
eth0: Setting half-duplex based on auto-negotiated partner ability 0000.
ehci-hcd 00:1d.7: resume
uhci-hcd 00:1d.2: resume
PCI: Setting latency timer of device 00:1d.2 to 64
uhci-hcd 00:1d.1: resume
PCI: Setting latency timer of device 00:1d.1 to 64
uhci-hcd 00:1d.0: resume
PCI: Setting latency timer of device 00:1d.0 to 64
Devices Resumed
Restarting tasks... done
init left refrigerator
[...]

I have no idea what is this device c03730cc.

With S3 sleeping state, things are more complicated. It worked sometimes
with the 2.5.63+acpi patch kernel I was using, but, most of the times,
the computer hung at suspend time at various places. I have not yet
managed to have it working with the current kernel. I will do some tests
and make a real bug report. But is it worth reporting S3 not working when
even S1 does not work ?

=C9ric Brunet