Re: 2.6.32 regression (bisected): Video tearing/glitching with T400laptops

[Jesse Barnes]

On Wed, 14 Oct 2009 22:26:45 -0400
Theodore Tso <tytso@xxxxxxx> wrote:

> On Wed, Oct 14, 2009 at 02:22:52PM -0700, Jesse Barnes wrote:
> > On Tue, 13 Oct 2009 12:14:26 -0700
> > Jesse Barnes <jbarnes@xxxxxxxxxxxxxxxx> wrote:
> > > According to the chipset team that form of display corruption is
> > > likely related to RAM self-refresh... Sounds like the display
> > > plane isn't getting its memory requests serviced fast enough when
> > > in self-refresh mode, which might mean we have to program the
> > > self-refresh watermarks more aggressively on GM45.
> > 
> > Ok, like any good bug there was more than one thing wrong:
> >   - we weren't setting up a fence for the object before enabling
> > FBC. Chris caught this and posted a patch to
> >     intel-gfx@xxxxxxxxxxxxxxxxxxxxx titled "drm/i915: Install a
> > fence register for fbc on g4x" (attached for convenience)
> >   - turns out we *do* need to set watermarks on G4x, despite some hw
> >     documentation indicating otherwise, patch for that attached
> > 
> > Hopefully with these two you'll have a solid display and some power
> > saving!
> 
> With these two patches (and none of your prior ones) applied to
> 2.6.32-rc3, things are definitely better.  I'm still seeing a few
> glitches (one as mutt was loading my Maildir inbox) and one as I was
> typing this message in emacs, but it's at best one minor every 5-10
> minutes.  It's rare enough to be at most a wee bit distracting (*lots*
> better than before), and I can see the power savings.
> 
> If I need to live with a display glitch every 5-10 minutes or so to
> get better power savings, I'll take it....

Oh I think we can do better than that.  The hardcoded values I put in
for the watermarks are almost certainly wrong.  I should be able to
reproduce what you're seeing (anything that takes a lot of memory
bandwidth will probably cause flicker) and adjust the watermarks to the
correct values to avoid that.  I'll try to find time in Japan to do
that...

-- 
Jesse Barnes, Intel Open Source Technology Center
--
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/