[clug] 2 questions (glut'ish)

[Ben Nizette]

On 20/08/2010, at 12:13 PM, Hal Ashburner wrote:

> Is this not a case where you should get support from nvidia seeing as you're their customer and have a problem with a bug in their product? Not very FOSS but hey, they might even fix it. Or you can let other people know to try and avoid them as a vendor?
> 

Certainly all the Player/Stage lists have been notified of the problem but with my thesis due in a month and this sim required to run the experiment to write about my focus was on getting a working solution fast!  Once again, not really the FOSS mentality but when I get a chance to breathe I'll be looking in to it (and a half-dozen other bugs I've uncovered).

	--Ben.

> On 20/Aug/10 12:01 PM, Ben Nizette wrote:
>> On 20/08/2010, at 10:22 AM, Andrew Janke wrote:
>> 
>>> Hi all,
>>> 
>>> 1. I am responsible for 2 legacy (~1990!) GLUT apps that were
>>> originally written on an SGI in GL and then I ported to OpenGL many
>>> years ago. Things have been rather peachy for a long time with only
>>> small changes required.
>>> 
>>> However there is currently a bit of a show stopper issue with a
>>> combination of the nvidia drivers in Ubuntu Lucid
>>> (173.14.22-0ubuntu11) and freeglut. I get colormap issues and the app
>>> runs but everthing is in black and white (1 bit? :). Everything is
>>> fine with the karmic nvidia binaries though...
>> I've been playing with the Player/Stage robotic framework and simulator recently, the sim renders using opengl.  It sometimes requests an invalid colour mode and every combination of graphics driver and distribution just prints an error and moves on except lucid+nvidia which starts corrupting memory and, if you're lucky, segfaults.
>> 
>> The invalid colour mode is deep in some library call so I can't fix it directly and of course I can't play with the driver so my "solution": 9.04 in a VM.
>> 
>> So yes, you aren't alone, but I'm afraid I can't be much more help than that (though let $list know how you go, thanks!) :-)
>> 
>> 	--Ben.
>> 
>>> So before I attempt to debug a binary blob, is anyone else out there
>>> using GLUT + nvidia and having the same issues?  As far as I can tell
>>> the bug is something to do with when the app requests the various
>>> display modes (glutGet(GLUT_DISPLAY_MODE_POSSIBLE)) and doesn't get
>>> what it expects, perhaps something has changed in the API...
>>> 
>>> 
>>> 
>>> 2. We should form a "linux party" for the next election.
>>> 
>>> 
>>> 
>>> a (1 list with wildly meandering threads is fine by me)
>>> -- 
>>> linux mailing list
>>> linux at lists.samba.org
>>> https://lists.samba.org/mailman/listinfo/linux
> 
> -- 
> linux mailing list
> linux at lists.samba.org
> https://lists.samba.org/mailman/listinfo/linux