Re: i915 regression in 2.6.29-rc5

From: Ingo Molnar
Date: Thu Feb 19 2009 - 06:27:34 EST



* Zhang Rui <rui.zhang@xxxxxxxxx> wrote:

> On Thu, 2009-02-19 at 03:33 +0800, Yinghai Lu wrote:
> > On Tue, Feb 17, 2009 at 7:24 PM, ZhangRui <rui.zhang@xxxxxxxxx> wrote:
> > > Hi, all,
> > >
> > > I got an i915 failure in 2.6.29-rc5.
> > >
> > > problem description: the screen doesn't come back after suspend/resume.
> > > A workaround is to boot with "nomodesetting".
> > >
> > > I got this error message during i915 driver load time,
> > > [ 22.606749] mtrr: type mismatch for e0000000,10000000 old: write-back new: write-combining
> > > [ 22.606752] [drm] MTRR allocation failed
> > > [ 22.606753] . Graphics performance may suffer.
> >
> > CONFIG_MTRR_SANITIZER=y
> > CONFIG_MTRR_SANITIZER_ENABLE_DEFAULT=1
> > CONFIG_MTRR_SANITIZER_SPARE_REG_NR_DEFAULT=1
> >
> > may help this one.
> >
> hah, it helps.
> the screen is back with this, while my previous kernel option is same
> except that ïCONFIG_MTRR_SANITIZER_ENABLE_DEFAULT=0.

Ok, great. I think we should flip the default of
MTRR_SANITIZER_ENABLE_DEFAULT from 0 to 1 in 2.6.30,
as we now seem to a better job of sorting out MTRRs
details than X does.

Yinghai, mind sending a patch for that?

Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/