Re: [PATCH 06/11] drm/i915: pass which operation triggered the frontbuffer tracking

From: Daniel Vetter <daniel@ffwll.ch>
To: Rodrigo Vivi <rodrigo.vivi@gmail.com>
Cc: intel-gfx <intel-gfx@lists.freedesktop.org>,
	Paulo Zanoni <paulo.r.zanoni@intel.com>
Subject: Re: [PATCH 06/11] drm/i915: pass which operation triggered the frontbuffer tracking
Date: Mon, 15 Dec 2014 09:43:35 +0100	[thread overview]
Message-ID: <20141215084335.GC27182@phenom.ffwll.local> (raw)
In-Reply-To: <CABVU7+tSN-98zNq64DZ=pd41Cv-HgKuVEAFeY5TeE+bDwyEpHg@mail.gmail.com>

On Fri, Dec 12, 2014 at 05:05:41PM -0800, Rodrigo Vivi wrote:
> On Mon, Dec 8, 2014 at 8:53 AM, Daniel Vetter <daniel@ffwll.ch> wrote:
> > On Mon, Dec 08, 2014 at 02:09:15PM -0200, Paulo Zanoni wrote:
> >> From: Paulo Zanoni <paulo.r.zanoni@intel.com>
> >>
> >> We want to port FBC to the frontbuffer tracking infrastructure, but
> >> for that we need to know what caused the object invalidation/flush so
> >> we can react accordingly: CPU mmaps need manual, GTT mmaps and
> >> flips don't need handling and ring rendering needs nukes.
> >>
> >> Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
> >> ---
> >>  drivers/gpu/drm/i915/i915_drv.h            |  7 +++++++
> >>  drivers/gpu/drm/i915/i915_gem.c            | 10 +++++-----
> >>  drivers/gpu/drm/i915/i915_gem_execbuffer.c |  2 +-
> >>  drivers/gpu/drm/i915/intel_drv.h           | 11 +++++++----
> >>  drivers/gpu/drm/i915/intel_frontbuffer.c   | 15 ++++++++++-----
> >>  5 files changed, 30 insertions(+), 15 deletions(-)
> >>
> >> diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
> >> index 9d694f1..ea3cc81 100644
> >> --- a/drivers/gpu/drm/i915/i915_drv.h
> >> +++ b/drivers/gpu/drm/i915/i915_drv.h
> >> @@ -685,6 +685,13 @@ struct intel_context {
> >>       struct list_head link;
> >>  };
> >>
> >> +enum fb_op_origin {
> >> +     ORIGIN_GTT,
> >> +     ORIGIN_CPU,
> >> +     ORIGIN_RENDER,
> >
> > Maybe GPU instead of RENDER since it includes the blitter? Render
> > typically only means the render ring in gem code. And maybe add an I915_
> > prefix or so at least to the enum.
> 
> I agree with Daniel that RENDER isn't a good name, but also not sure
> about GPU... Maybe ENGINE_CS or RINGS...

ORIGIN_CS sounds good to me. ENGINE is a bit misleading since it's for all
engines - usually we use engine only when we mean a specific one. Paulo,
can you please pick one of the color suggestions and then please resend?
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx