Re: another data point (was Re: acpi_vbios_reset hangs while vbetool works (Was: sporadic failure to wakeup from S3))

[David Brownlee <abs%netbsd.org@localhost>]

2011/9/8 Marko Schütz Schmuck <MarkoSchuetz%web.de@localhost>:
> At Tue, 6 Sep 2011 18:25:32 +0100,
> David Brownlee wrote:
>>
>> 2011/9/5 Marko Schütz Schmuck <MarkoSchuetz%web.de@localhost>:
>> > At Mon, 05 Sep 2011 15:14:02 -0400,
>> > Marko Schütz Schmuck wrote:
>> >>
>> >> At Mon, 05 Sep 2011 15:06:26 -0400,
>> >> Marko Schtz Schmuck wrote:
>> >> >
>> >> > Recently, I am experiencing sporadic failure to wake from ACPI Sleep
>> >> > state S3.
>> >
>> > now using machdep.acpi_vbios_reset=0 together with appropriate vbetool
>> > invocations in /etc/powerd/scripts/sleep_button and I haven't had the
>> > failure to wake since, but it's only been a short time...
>> >
>> > When using 5.0.2 I never had to set machdep.acpi_vbios_reset=0 nor did
>> > I have to use vbetool.
>
> the problem persists: I still have sporadic failure to wake from
> S3. It does not seem to make a difference whether I use
>
> machdep.acpi_vbios_reset=0
>
> with vbetool or
>
> machdep.acpi_vbios_reset=1
>
> I'm wondering whether the problem might lie with some other part of
> the wake-up other than vbios. Any ideas?
>

I assume you have exactly the same hardware connected.
If you switch away from X before suspend/resume does it affect
matters, similarly if its on battery/mains.

It could be a race condition somewhere, which are almost always a huge
quantity of fun (FSVO fun).