[Mesa-dev] EGL/android: pbuffer implementation.

Xu, Randy randy.xu at intel.com
Fri Nov 11 01:22:54 UTC 2016


Hi, Emil

I may misunderstand your meaning, you want to Zhiquan to refine the code and avoid duplicating those identical codes, right?
In this case, I agree with you fully

Thanks,
Randy

-----Original Message-----
From: Xu, Randy 
Sent: Friday, November 11, 2016 9:01 AM
To: Emil Velikov <emil.l.velikov at gmail.com>; Liu, Zhiquan <zhiquan.liu at intel.com>
Cc: ML mesa-dev <mesa-dev at lists.freedesktop.org>; Long, Zhifang <zhifang.long at intel.com>; Kondapally, Kalyan <kalyan.kondapally at intel.com>; Xu, Randy <randy.xu at intel.com>
Subject: RE: [Mesa-dev] EGL/android: pbuffer implementation.

Hi, Emil

The commit 63c5d5c6c46c8472ee7a8241a0f80f13d79cb8cd implements the pbuffer for surfaceless platform, while Zhiquan's patch is for Android platform. They are similar indeed, but we need both. This patch can solve 300+ dEQP failure on Android platform due to pbuffer crash.

Thanks,
Randy

-----Original Message-----
From: Emil Velikov [mailto:emil.l.velikov at gmail.com]
Sent: Friday, November 11, 2016 1:02 AM
To: Liu, Zhiquan <zhiquan.liu at intel.com>
Cc: ML mesa-dev <mesa-dev at lists.freedesktop.org>; Long, Zhifang <zhifang.long at intel.com>; Kondapally, Kalyan <kalyan.kondapally at intel.com>; Xu, Randy <randy.xu at intel.com>
Subject: Re: [Mesa-dev] EGL/android: pbuffer implementation.

On 10 November 2016 at 11:48, Liu Zhiquan <zhiquan.liu at intel.com> wrote:
> mesa android path didn't support pbuffer, so add pbuffer support to 
> fix most deqp and cts pbuffer test cases fail; add support of front 
> buffer and single buffer config.
>
> Test status: android CTS EGL pbuffer test can run without native crash.
> test:[DEQP,EGL]all deqp pbuffer case passed.
>
I'm leaning that a lot of the reasoning behind commit 63c5d5c6c46c8472ee7a8241a0f80f13d79cb8cd and the associated discussion [1] should be applicable in this case ?

Temporary ignoring all those: the patch duplicates an identical code - once for the front and once for the back buffer.
Not cool ...

-Emil
[1] https://patchwork.freedesktop.org/patch/85913/


More information about the mesa-dev mailing list