search

paulpairbus / glintercept

Automatically exported from code.google.com/p/glintercept
0 stars 0 forks source link

GLIntercept shows many false negatives #7

Open GoogleCodeExporter opened 9 years ago

GoogleCodeExporter commented 9 years ago 
I ran a program where I added my OpenGL code over a given infrastructure.

In GLIntercept error log I see a lot of "glEnd with no previous glBegin calls" 
messages, but in the full log when I track it down I see that each glBegin has 
its glEnd matched perfectly.

Same goes for glNewList and glEndList.

Original issue reported on code.google.com by ita...@gmail.com on 12 Nov 2012 at 11:26

GoogleCodeExporter commented 9 years ago 
Can you post a log trace of GLIntercept function calls that show this issue? 
I would be very surprised if there is a bug in GLIntercept.

Are you making use of multiple contexts? Multiple threads? Have you tried 
turning on the thread checking?

Is there a sample program I can download to check this myself?

Original comment by dtrebi...@gmail.com on 12 Nov 2012 at 10:50

GoogleCodeExporter commented 9 years ago 
I'm not sure about having different openGL contexts. I'm not creating any 
context but drawing on a given context (I'm using a dynamic layer of Esri 
ArcMap in immediate mode), so it is possible that they do such openGL context 
switching.
I couldn't reproduce it with a pure openGL program of my own.

I've attached a zip with the GLIntercept logs showing the false negatives.
If you filter just the glBegin and glEnd calls you can see they all match, so 
the errors are unclear.

It would be helpful to place match markers on both files so one can tell for 
each error where in the full-log should the error be looked for.

Original comment by ita...@gmail.com on 13 Nov 2012 at 8:16

Attachments:

GoogleCodeExporter commented 9 years ago 
What is the gliConfig.ini you used to generate this log?

Original comment by dtrebi...@gmail.com on 13 Nov 2012 at 1:18

GoogleCodeExporter commented 9 years ago 
The default one.

Original comment by ita...@gmail.com on 13 Nov 2012 at 1:18

GoogleCodeExporter commented 9 years ago 
Try the FullDebug one. (or turn on all the options in the ErrorChecking section)

Original comment by dtrebi...@gmail.com on 13 Nov 2012 at 2:02

GoogleCodeExporter commented 9 years ago 
Just did. Same problem.

Original comment by ita...@gmail.com on 13 Nov 2012 at 2:15

GoogleCodeExporter commented 9 years ago 
I suspect I will have to debug what is going on - is there any way you can give 
me a sample bit of code that demostrates this issue? (I might be able to get a 
copy of ArcGIS to test)

Original comment by dtrebi...@gmail.com on 13 Nov 2012 at 9:56

GoogleCodeExporter commented 9 years ago 
If you'll get a copy of ArcGIS (I use the ArcEngine 9.3.1 SP 1) I'll send you a 
sample program.

Original comment by ita...@gmail.com on 14 Nov 2012 at 5:51

GoogleCodeExporter commented 9 years ago 
Will a trial version of the ArcGIS desktop work? or do I need something else?

Original comment by dtrebi...@gmail.com on 14 Nov 2012 at 6:21

GoogleCodeExporter commented 9 years ago 
It could be OK as long as it's the version I specified above (9.3.1 SP1).
But you'll probably be only able to get version 10 or 10.1 now (make sure you 
get the "engine", not the "runtime"). For that version I also have a sample 
program, but I never tested it with GLIntercept (and since it's not installable 
side-by-side with V9.3.1 I can't check it again now). I ended up not using the 
newer version because it didn't work for me.

Original comment by ita...@gmail.com on 14 Nov 2012 at 8:14