[cairo] Cairo OpenGL ES 2.0 backend is rotting away?

[Chris Wilson]

On Mon, 23 Apr 2012 17:15:31 +0300, Alexandros Frantzis <alexandros.frantzis at linaro.org> wrote:
> On Mon, Apr 23, 2012 at 03:39:28PM +0300, Siarhei Siamashka wrote:
> > Hello,
> > 
> > Tried cairo-perf-trace benchmark using GLES2 backend (--enable-glesv2) on
> > x86-64 computer with open source radeon drivers.
> > 
> > === after the merge of GLES2 in 4d96859ba5eb6018ae257ef6bfead228583908cf ===
> > 
> > [ # ]  backend                         test   min(s) median(s) stddev. count
> > [  0]      egl             firefox-fishtank   10.282   10.282   0.00%    1/1
> > 
> > === starting with 2855ff4666922f2c38505414270d47f659b0d499 ===
> > 
> > [ # ]  backend                         test   min(s) median(s) stddev. count
> > [  0]      egl             firefox-fishtank    0.000    0.000   -nan%    0/1
> > 
> > === starting with af9fbd176b145f042408ef5391eef2a51d7531f8 ===
> > 
> > Segmentation fault
> > 
> > === current git master abedc6b46ad283f896e078479174312cb87e6700 ===
> > 
> > cairo-perf-trace: cairo-gl-operand.c:450: _cairo_gl_operand_destroy:
> > Assertion `!"reached"' failed.
> > 
> > -- 
> > Best regards,
> > Siarhei Siamashka
> > 
> 
> Hi,
> 
> the patch below fixes things for me (tried it on a x86-64 with r600g).
> Please try it out and check if it fixes the perf test case in your system, too.

Thank you Alexandros for the patch, and Siarhei for testing.
Pushed to master,
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre