[Mesa-dev] [RFC 00/12] egl, i965: Support EGL_ANDROID_native_fence_sync (v2)

[Rob Clark]

On Thu, Oct 13, 2016 at 4:37 PM, Rafael Antognolli
<rafael.antognolli at intel.com> wrote:
> Hi all,
>
> On Tue, Oct 11, 2016 at 01:47:37PM -0400, Robert Foss wrote:
>>
>>
>> On 2016-10-11 01:37 PM, Chad Versace wrote:
>> > On Tue 11 Oct 2016, Emil Velikov wrote:
>> > > On 10 October 2016 at 20:41, Rob Clark <robdclark at gmail.com> wrote:
>> > > > On Mon, Oct 10, 2016 at 1:43 PM, Chad Versace <chadversary at chromium.org> wrote:
>> >
>> > > > > Testing:
>> > > > >
>> > > > >     I haven't yet tested EGL_ANDROID_native_fence_sync with these patches, but
>> > > > >     the series does build. I just wanted to get these patches on the list as a
>> > > > >     status-update.  I'm slowly working on getting Rob Clark's hacked-up kmscube to
>> > > > >     work on this series.
>> > > > >
>> > > > >     I did test that this series doesn't break any dEQP-EGL sync tests. But my
>> > > > >     dEQP doesn't include any EGL_ANDROID_native_fence_sync tests.
>> > > > >
>> >
>> > > From a quick skim neither one of CTS/dEQP/piglit had
>> > > EGL_ANDROID_native_fence_sync tests. I've mentioned to Rob F that
>> > > adding a few piglits might be a good idea. It would be good to
>> > > coordinate to avoid writing the same tests twice ;-)
>> >
>> > Yes, we should have tests before committing this feature. I am releasing
>> > any lock I may had held on test writing, as I don't have time to tackle
>> > that this week.
>>
>> Writing the tests is not on my table at the moment. Perhaps some time can be
>> scraped together for it at a later point, but not now.
>
> If nobody has started writing piglit tests yet, and it's OK with you
> guys, I would like to give it a try since I'm also interested in this
> feature.

that would be hugely useful.. I'm not sure anyone else has started on
tests, but they are definitely needed

> Also if I understood correctly, the initial thing to test is whether
> WaitSync and ClientWaitSync respect input fences, right? I'll try to
> come up with a list of other things to test, like invalid fences and so
> on, but if you guys already have a list of things to test in mind,
> please let me know.

I guess API level tests, like can you create a fence w/
NO_NATIVE_FENCE_FD as input, then glFlush(), and then get an fd back
from eglDupFenceFD(), and that sort of thing?  I haven't really given
it much thought yet.

BR,
-R

> Anyway, is it fine if I start working on this?
>
> Regards,
> Rafael