RESOLVED FIXED 146526
Disable experimental WebGL2 implementation 
https://bugs.webkit.org/show_bug.cgi?id=146526
Summary Disable experimental WebGL2 implementation
Dean Jackson
Reported 2015-07-01 15:13:49 PDT
Disable WebGL2
Attachments
Patch (44.18 KB, patch)
2015-07-01 15:26 PDT, Dean Jackson 		mmaxfield: review+
DetailsFormatted DiffDiff
View All Add attachment proposed patch, testcase, etc.
Dean Jackson
Comment 1 2015-07-01 15:14:45 PDT
Our experimental implementation is not ready to be exposed.
Radar WebKit Bug Importer
Comment 2 2015-07-01 15:15:34 PDT
<rdar://problem/21641229>
Radar WebKit Bug Importer
Comment 3 2015-07-01 15:15:37 PDT
<rdar://problem/21641235>
Dean Jackson
Comment 4 2015-07-01 15:19:42 PDT
Not sure why we got two radars. Which one updates?
Dean Jackson
Comment 5 2015-07-01 15:26:19 PDT
Created attachment 255957 [details] Patch
Myles C. Maxfield
Comment 6 2015-07-01 15:31:56 PDT
Comment on attachment 255957 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=255957&action=review > Source/WTF/wtf/FeatureDefines.h:766 > +#if !defined(ENABLE_WEBGL2) > +#define ENABLE_WEBGL2 0 > +#endif Why is this necessary? Weinig says we should be trying to pare down this file, not add to it. > Source/WebCore/bindings/js/JSCanvasRenderingContextCustom.cpp:42 > +#if ENABLE(WEBGL2) > +#include "JSWebGL2RenderingContext.h" > +#endif Don't we usually have a pattern to put these ENABLE() guards inside the header files themselves?
Myles C. Maxfield
Comment 7 2015-07-01 15:32:04 PDT
rs=me
Dean Jackson
Comment 8 2015-07-01 15:37:27 PDT
Comment on attachment 255957 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=255957&action=review >> Source/WTF/wtf/FeatureDefines.h:766 >> +#endif > > Why is this necessary? Weinig says we should be trying to pare down this file, not add to it. I agree! I was the one who proposed removing this file :) But until there is a plan, I would like to keep consistent. >> Source/WebCore/bindings/js/JSCanvasRenderingContextCustom.cpp:42 >> +#endif > > Don't we usually have a pattern to put these ENABLE() guards inside the header files themselves? Good catch. It's actually automatically in that file thanks to the Condition guard.
Dean Jackson
Comment 9 2015-07-01 15:40:03 PDT
Committed r186198: <http://trac.webkit.org/changeset/186198>
Chris Dumez
Comment 10 2015-07-01 17:32:37 PDT
Look like this needs rebaselining: js/dom/global-constructors-attributes.html
Chris Dumez
Comment 11 2015-07-01 17:42:04 PDT
(In reply to comment #10) > Look like this needs rebaselining: > js/dom/global-constructors-attributes.html <http://trac.webkit.org/changeset/186211>
Note You need to log in before you can comment on or make changes to this bug.