631643 - Check for ARB_ES2_compatibility before using it, fix MAX_VARYING_VECTORS emulation, clean up getParameters

Status: RESOLVED FIXED

(Core :: Graphics: CanvasWebGL, defect)

Description

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

Attachment: patch

This should make us pass a few tests, so need a tryserver run and update failing-tests-xx.txt files.

Comment 1

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

OK, this makes us pass gl-get-calls.html

Comment 2

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

http://hg.mozilla.org/mozilla-central/rev/926ea5719563

Comment 3

[Kyle Huey (Exited; not receiving bugmail, old account, do not use)]

Backed out due to test failures

http://hg.mozilla.org/mozilla-central/pushloghtml?changeset=80211f053c46

Comment 4

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

So, I made a new tryserver run with lots of extra debug info,

http://tinderbox.mozilla.org/showlog.cgi?log=MozillaTry/1297554321.1297556558.29898.gz#err0

and it tells me that on Mac, the extension ARB_ES2_compatibility is absent (as it is on Linux), we then call glGetIntegerv(LOCAL_GL_MAX_VERTEX_OUTPUT_COMPONENT) and that generates a GL_INVALID_ENUM error on Mac, while it doesn't generate an error on Linux.

Comment 5

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

OK, here's the explanation: MAX_VERTEX_OUTPUT_COMPONENT is mentioned for the first time in the OpenGL 3.2 spec.

Comment 6

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

Attachment: patch updated

New patch takes a simpler way to computing mGLMaxVaryingVectors: instead of querying these opengl 3.2 pnames, just use their minimum legal value from the spec.

Pushed to try: http://tbpl.mozilla.org/?tree=MozillaTry&rev=9a7f6ef1060a

Comment 7

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

Comment on attachment 512035 [details] [diff] [review]
patch updated

Hmm, I don't really like this; there must be a way to query this earlier?  Otherwise we should just skip querying if we're on GL < 3.2 (or don't have whatever extension they came in on).  We already don't give authors all that much information about the hardware they're on, let's not lie to them about capabilities...

Comment 8

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

Attachment: patch updated

OK, here you go. Now we try to query these values, and carefully glGetError() to handle the case where they're not available.

Comment 9

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

Comment on attachment 512205 [details] [diff] [review]
patch updated

Looks great, thanks!

Comment 10

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

http://hg.mozilla.org/mozilla-central/rev/0dac834eae79