Compilation error with GLES2 when the gl2.h driver is of an older version.
Created attachment 151808 [details] Patch
Comment on attachment 151808 [details] Patch Clearing flags for now.
Created attachment 151825 [details] Patch
Created attachment 151827 [details] Patch
Comment on attachment 151827 [details] Patch Looks like the best available solution at this point. r=me
Comment on attachment 151827 [details] Patch Clearing flags on attachment: 151827 Committed r122413: <http://trac.webkit.org/changeset/122413>
All reviewed patches have been landed. Closing bug.
This breaks the BlackBerry build, so I am trying to find a solution. We have a version of gl2ext.h from 2010 which defines the problematic symbols with -PROC. The newest version from the Khronos website (http://www.khronos.org/registry/gles/api/2.0/gl2ext.h) also defines the symbols in conflict with -PROC. The ANGLE header, which according to the revision statement is newer than ours but older than Khronos' defines the symbols in conflict without -PROC. Does this conflict only affect BlackBerry? Are any of the headers wrong?
(In reply to comment #8) > This breaks the BlackBerry build, so I am trying to find a solution. > > We have a version of gl2ext.h from 2010 which defines the problematic symbols with -PROC. The newest version from the Khronos website (http://www.khronos.org/registry/gles/api/2.0/gl2ext.h) also defines the symbols in conflict with -PROC. The ANGLE header, which according to the revision statement is newer than ours but older than Khronos' defines the symbols in conflict without -PROC. The Khronos gl2ext.h header was just updated to add the -PROC suffix to these symbols based on a bug I filed. Before that, the symbols were defined without the PROC suffix for quite some time -- I'm pretty sure since before 2010. > Does this conflict only affect BlackBerry? Are any of the headers wrong? Yes, the conflict currently only affects Blackberry. You should add an #ifdef to the WebCore code to get things compiling.