705002 - Fix OGL Layers for Windows

Status: RESOLVED FIXED

(Core :: Graphics, defect)

Description

[Kelsey Gilbert [:jgilbert]]

To my knowledge, there's no reason OGL Layers should fail to work properly on windows, since it should be readily cross-platform. I don't think we're likely to ever use it by default, since our D3D backend is much more reliable for most Windows machines, but it's a little worrisome that OGL Layers isn't working at all on Windows.

Comment 3

[Vladimir Vukicevic [:vlad] [:vladv] (needinfo me, slow to respond)]

Attachment: small debug build fix

Seems to work fine; needs a small fix for debug builds, but otherwise comes up just fine.

Comment 4

[Benoit Jacob [:bjacob] (mostly away)]

Comment on attachment 765971 [details] [diff] [review]
small debug build fix

Review of attachment 765971 [details] [diff] [review]:
-----------------------------------------------------------------

::: gfx/gl/GLContext.cpp
@@ +926,5 @@
>      MOZ_ASSERT(caps.alpha == !!format.alpha);
> +
> +    // These we either must have if they're requested, or
> +    // we can have if they're not.
> +    MOZ_ASSERT(caps.depth == !!format.depth || !caps.depth);

This doesn't seem to do what the comment says: if caps.depth==false, and format.depth==0, the old assert used to pass, and the new one fails.

Taking a step back, there is no chance that a boolean formula in 2 variables would need to mention one of the variables twice. The assert condition as currently written simplifies to:

MOZ_ASSERT(!!format.depth && caps.depth);

Comment 5

[Benoit Jacob [:bjacob] (mostly away)]

Comment on attachment 765971 [details] [diff] [review]
small debug build fix

Oh, sorry. Operator priority is hard. On fridays.

Comment 6

[Vladimir Vukicevic [:vlad] [:vladv] (needinfo me, slow to respond)]

 https://hg.mozilla.org/integration/mozilla-inbound/rev/1f72257122b5

Comment 7

[Kelsey Gilbert [:jgilbert]]

Typo in the bug number, fwiw.

Also, should we add a reftest (for webgl) for this? This would give us at least non-zero coverage.

Comment 8

[Vladimir Vukicevic [:vlad] [:vladv] (needinfo me, slow to respond)]

pretty sure this got merged to m-c (but wrong bug #)

Comment 9

[Kelsey Gilbert [:jgilbert]]

Yep:
https://hg.mozilla.org/mozilla-central/rev/1f72257122b5