[Intel-gfx] [PATCH 2/8] drm/i915: skip redundant operations whilst enabling pipes and planes

[Jesse Barnes]

On Fri, 18 Mar 2011 17:37:06 +0000
Chris Wilson <chris at chris-wilson.co.uk> wrote:

> On Fri, 18 Mar 2011 09:25:22 -0700, Jesse Barnes <jbarnes at virtuousgeek.org> wrote:
> > On Fri, 18 Mar 2011 08:02:04 +0000
> > Chris Wilson <chris at chris-wilson.co.uk> wrote:
> > 
> > > If the pipe or plane is already enabled, then we do not need to enable
> > > it again and can skip the delay.
> > > 
> > > References: https://bugs.freedesktop.org/show_bug.cgi?id=34601
> > > Cc: Jesse Barnes <jbarnes at virtuousgeek.org>
> > > Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> > > ---
> > 
> > Does this one actually fix the issue?  Unless we're trying to get
> > suspend, resume, or mode setting down under 100ms it's probably not
> > worth it to try to avoid the vblank waits (this is me being paranoid
> > about our mode setting code).
> 
> This is the one QA identified as fixing the regression from b24e717988,
> which worries me.
> 
> I admit to not knowing what the root cause is, but it was a change
> introduced by b24e71798 and had unfortunate side-effects.

Yeah that message confused me; they also identified the other two fixes
working (both the vblank wait on gen2 and the display plane flush).
I've asked for clarification.

-- 
Jesse Barnes, Intel Open Source Technology Center