[Mesa-dev] [PATCH v1 0/7] Implement commont gralloc_handle_t in libdrm

Tapani Pälli tapani.palli at intel.com
Thu Mar 22 13:06:55 UTC 2018



On 03/22/2018 01:01 PM, Emil Velikov wrote:
> Hi guys,
> 
> On 22 March 2018 at 08:23, Tomasz Figa <tfiga at chromium.org> wrote:
>> Hi Chih-Wei,
>>
>> On Thu, Feb 22, 2018 at 2:53 PM, Chih-Wei Huang <cwhuang at linux.org.tw> wrote:
>>> 2018-02-21 3:03 GMT+08:00 Rob Herring <robh at kernel.org>:
>>>>
>>>> Perhaps worth revisiting. Given we've failed to progress at all since
>>>> then may change opinions some. We already have to handle multiple
>>>> opens share the same pipe_screen, so I don't think reusing the fd buys
>>>> us anything.
>>>>
>>>> Maybe we're close to the point of removing the flink name support too.
>>>> The android-x86 folks have been working to get dma-bufs working.
>>>> Chih-Wei, any comments on this?
>>>
>>> Ah! Sorry. I didn't catch or understand the details.
>>> Did you mean the attempts to use drm_hwcomposer
>>> in android-x86?
>>> My understanding so far is most x86 GPUs won't work
>>> except some very limited models.
>>> It's due to kernel driver issues which may never be solved.
>>> So we can't drop the flink name support.
>>> Please correct me if I am wrong.
>>
>> Could you elaborate a bit more on those GPUs that won't work and
>> corresponding driver issues? We're running cros_gralloc on Intel and
>> AMD GPUs, with DMA-buf and render-node only setup and we haven't seen
>> any problems.
>>
> May I suggest an alternative to, diving into details of the issues
> observed by the Android-x86 team.
> 
> Namely: Tomasz, please send a patch ripping all the flink stuff, thus
> Chih-Wei and others can test it.
> 
> Ideally it'll be a quick and easy "works for me" ;-)

This has been discussed before and IIRC one reason is that i915 does not 
have dmabuf implemented.

Overall there seems to be more dmabuf users than flink users. Therefore 
it would make sense to default to dmabuf and flink users should carry 
patches instead.

// Tapani


More information about the mesa-dev mailing list