The userguide says that there is a leak on Windows XP on Mobility Radeon 9700 hardware.
The bug is documented in JOGL Issue 166, which is marked as RESOLVED. But it was marked as RESOLVED only 41 minutes after it was opened.
So is the issue really resolved? If so, then the userguide should be updated. If not, then the issue should be reopened.
I don’t have a confirmed report that ATI has fixed this bug in their drivers. The code that was put into JOGL back in 2005 as per the bug report was a workaround, not a fix. If you have this kind of hardware and are running ATI’s latest drivers, please adapt the test case to the new javax.media.opengl.* APIs and report whether the memory leak has since been fixed. If so, I’ll update the user guide.
As far as I can tell, no code was put into JOGL back in 2005 as per the bug report. The bug report just says that end-users could add -Djogl.GLContext.nofree to the Java command line to work around the bug.
Unfortunately I do not have this kind of hardware, so someone else will have to test it.
The bug report says “Worked around memory leak in ATI’s OpenGL drivers by adding system property -Djogl.GLContext.nofree which users can specify on the command line.” Code was added to JOGL at this point (by me) and this command line option is still supported.