Using VESA to restore display settings on resume

[Vladimir Grebenschikov]

В вс, 27/02/2005 в 23:26 -0500, Josh Elsasser пишет:
...cut...

>> >> Patch also at: http://people.freebsd.org/~iedowse/vesa_restore.diff
>> >> 
>> >
>> >THANK YOU!  I've had my laptop for 3 1/2 years, and I can finally
>> >suspend/resume and have a usable display afterwards!  Cardbus doesn't
>> >appear to work after resume, but that's a whole different can of
>> >worms...
>> >
>> >The system I tested this on was RELENG_5 from yesterday, the patch
>> >applied cleanly.  I already had the hw.acpi.reset_video set to 0, so I
>> >don't know if it would work with it set to 1.
>> 
>> What notebook you have ? I was not such lucky.
>> 
>> I Has SONY VAIO PCG-Z1WA, by zzz it sleeps ok, 
>> With patch it even said about saving VESA data.
>> On any key after it is awakes, spin up drive and fan, but screen still
>> black (even no backlight), and it is very unlike that I get console back
>> (typing has no effect, even when I try to call kdb).
>> 
>> Also if I try this patch with vesa module loaded it just carsh in vm86
>> mode on zzz (curproc acpiconf).
>> 
>> Any suggestions are very welcome.
>
>What is hw.acpi.reset_video set to on your system?  At some point
>between 5.2 and 5.3, my machine would hang on resume instead of just
>resuming with no display.  Setting hw.acpi.reset_video=0 in
>loader.conf fixed that.  I have a Sony Vaio FX220, which has an i815
>graphics chip.
>
I've tried both 0 and 1 for reset_video (but set via sysctl).
And I has 6-CURRENT now, but this notebook never returned from suspend 
(with 5.3 too). It has ATI RADEON chip.

Probably I should try firewire console to detect what happens on resume,
but firewire bus itself should be resumed first. 

Suspend/Resume gurus, is it possible to debug dead console resume with
firewire ?

> -jre

-- 
Vladimir B. Grebenchikov
vova at fbsd.ru