gpu: Prefer GL_ARB_sync over GL_NV_fence.

gpu: Prefer GL_ARB_sync over GL_NV_fence.

GL_ARB_sync allows for acquiring fences on the server side.

NOTRY=True

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=257163

[no sievers, 2014-03-13 19:56:45]

Depends on https://codereview.chromium.org/180723023/

[piman, 2014-03-13 20:39:50]

On 2014/03/13 19:56:45, sievers wrote:
> Depends on https://codereview.chromium.org/180723023/

LGTM in principle, but why not fold it into 180723023?

[no sievers, 2014-03-13 20:53:55]

On 2014/03/13 20:39:50, piman wrote:
> On 2014/03/13 19:56:45, sievers wrote:
> > Depends on https://codereview.chromium.org/180723023/
> 
> LGTM in principle, but why not fold it into 180723023?

I'm just paranoid that using a different API introduces a regression somewhere
:)

I've seen this before on one driver where both GL_NV_fence and EGL fences were
exposes. The one API had a bug in testing fences, the other did not.

[no sievers, 2014-03-14 17:34:02]

The CQ bit was checked by sievers@chromium.org

[commit-bot: I haz the power, 2014-03-14 17:34:44]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/sievers@chromium.org/195763015/1

[commit-bot: I haz the power, 2014-03-14 17:37:30]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-03-14 17:37:31]

Try jobs failed on following builders:
  tryserver.chromium on linux_chromium_clang_dbg

[no sievers, 2014-03-14 19:01:07]

The CQ bit was checked by sievers@chromium.org

[commit-bot: I haz the power, 2014-03-14 19:01:17]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/sievers@chromium.org/195763015/1

[commit-bot: I haz the power, 2014-03-14 19:03:41]

Change committed as 257163