D23881: Possible solution for scheduling repaints on NVIDIA

[Fredrik Höglund]

fredrik marked an inline comment as done.
fredrik added a comment.


  For the record, I wouldn't mind if someone who is in a better position to test it commandeered this revision. I think I mentioned that on irc.

INLINE COMMENTS

> romangg wrote in glxbackend.cpp:301
> Can there be a case where we don't have intel-swap-event but also don't want the swap-handler to be activated? Some legacy systems for example.
> 
> I remember you provided the table with which hardware uses what extension so you know I'm sure.

The swap-handler should work with other implementations (famous last words?), but we don't know if SwapBuffers() is going to block. If it doesn't, we can't rely on the swap event for scheduling. So if we want to play it safe, we should only enable it on NVIDIA.

> romangg wrote in glxbackend.cpp:119
> Ok, let's remember this information. We can first test on master without it though.

It's not going to work properly until that has been fixed, so we certainly can't push it to master before then.

I think the easiest way to fix it is to save the attribute builder or the attribute list that was used to create the main context, and reuse it.

REPOSITORY
  R108 KWin

REVISION DETAIL
  https://phabricator.kde.org/D23881

To: fredrik, #kwin, romangg
Cc: davidedmundson, ekurzinger, kwin, LeGast00n, The-Feren-OS-Dev, sbergeron, jraleigh, fbampaloukas, GB_2, mkulinski, ragreen, jackyalcine, iodelay, crozbo, bwowk, ZrenBot, ngraham, alexeymin, himcesjf, lesliezhai, ali-mohamed, hardening, romangg, jensreuterberg, abetts, sebas, apol, ahiemstra, mart
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kwin/attachments/20191212/5e2ecfc3/attachment.html>