Snow Leopard help

I'm totally new to JOGL so it's possible this is a newbie mistake.

I downloaded the latest build (2.1.0b23 for OS X), uncompressed it, cd'd to the root of the unzipped packege and ran etc/test.sh

The result was:

Invalid memory access of location 0x25 rip=0x7fff8000933c

shortly after a small, grey window popped up.

Theorising that this might be due to some 32 vs 64 bit issue in the native libraries, I modified test.sh to include the -d32 java command line option. The result was not materially different:

Invalid memory access of location 0x28 eip=0x92510ed7

Is this some newbie mistake, or a problem with the build? I've uploaded my test.log just in case it proves useful.

test.log

Many thanks in advance for any advice or info

Jon

I don't think you're doing anything wrong. I saw someone on another thread mention this same problem a few days ago. It sounds like test.sh is calling something at the end that's not supported on Mac yet, but JOGL2 may still work for your program. You can see that it's found the GL extensions of your card, at least.

You might try one of the simple test programs at http://jogamp.org/wiki/index.php/Using_JOGL_in_AWT_SWT_and_Swing and see if it works.

Thanks, Wade.

Very useful examples, thanks!

I've made some progress and got some simple OpenGL rendering. Bit concerned that I'm eventually going to run into whatever was causing test.sh to crash...

I also cut/paste the SWT example you indicated and ran it. Works fine... with one exception - the application does not terminate after the window closes (although the event loop seems to terminate and the display.dispose() at the end is reached). I get the same problem with my own SWT app. I should probably post this in a separate thread, but thought I'd mention it since it applies to the examples you wrote.

Cheers
Jon

Further investigation shows that sticking a System.exit(0) after the display.dispose() fixes the problem for me in the SWT example. Is it normal to have to do this?

jonessex wrote

Further investigation shows that sticking a System.exit(0) after the display.dispose() fixes the problem for me in the SWT example. Is it normal to have to do this?

No, this is not normal  I've seen this inability to exit in other cases as well, and I've got a bug in on it at https://jogamp.org/bugzilla/show_bug.cgi?id=451 (for the Linux case, at least). You can enter a bug on it for SWT if you want, or I'll do it once I have time for some more testing on the Mac.