[Intel-gfx] About KMS, FB and Intel drivers

[Jesse Barnes]

On Wednesday, January 21, 2009 1:52 pm Yves-Alexis Perez wrote:
> On mer, 2009-01-21 at 12:54 -0800, Jesse Barnes wrote:
> > On Wednesday, January 21, 2009 12:37 pm Yves-Alexis Perez wrote:
> > > On Wed, Jan 21, 2009 at 09:04:35AM -0800, Jesse Barnes wrote:
> > > > I just pushed the fix, so you should be able to pull and get it.
> > >
> > > Thanks, I just pulled it and built it. Seems to work fine. I had a hang
> > > (xfwm was stuck, cursor moved fine etc but no way to change anything),
> > > which I couldn't reproduce. Maybe it was related to the fact I ran it
> > > through gdm (direct from console or running from slim it seems to work
> > > fine)
> > >
> > > I still have the WARNING kernel problem, and suspend fails. Well,
> > > suspend does success, but as soon as the laptop is suspended, it tries
> > > to resume and when the screen is on (with a blinking console cursor)
> > > (huge one, from a 80×24 console), the little led moon on the thinkpad
> > > is still on, and I can't do anything.
> >
> > The warning should be fixed in airlied's tree.  Suspend/resume has been
> > somewhat flakey for me with 2.6.29-rc* in general, but be sure you're not
> > running a POST at resume time (no s3_bios options on the command line, no
> > vbetool post in your resume path).
>
> Will try airlied tree. Master branch? drm-next? gem-updates? I don't

His drm-fixes branch has one patch to avoid the warnings.

> think I'm using vbetool, but I suspend using pm-utils and the quirk for
> my Thinkpad are:
>
> <merge key="power_management.quirk.s3_bios" type="bool">true</merge>
> <merge key="power_management.quirk.s3_mode" type="bool">true</merge>
>
> So it's worth trying without.

Yeah.

-- 
Jesse Barnes, Intel Open Source Technology Center