[Glamor] [PATCH 1/9 v2] glamor-composite: Optimize the computation of composite

Chris Wilson chris at chris-wilson.co.uk
Mon Jan 9 05:27:57 PST 2012


On Mon,  9 Jan 2012 21:01:10 +0800, zhigang.gong at linux.intel.com wrote:
> From 03273d2534622665e42c875b1fdc7e26f25838e5 Mon Sep 17 00:00:00 2001
> From: Zhigang Gong <zhigang.gong at linux.intel.com>
> Date: Sun, 8 Jan 2012 01:14:12 +0800
> Subject: [PATCH 1/9 v2] glamor-composite: Optimize the computation of composite
>  region.
> 
> Computing the composite region at the composite_with_shader is very
> inefficient. As when we call to here from the glamor_glyph's temproary
> picture, we don't need to compute this region at all. So we move this
> computing out from this function and do that at the glamor_composite
> function. This can get about 5% performance gain for aa10text/rgb10text.

Right, that's the bug I hit, not populating prects. It still doesn't
handle the malloc failure, you could just fail the op entirely, or 
submit 10 boxes at a time.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre


More information about the Glamor mailing list