Date: Wed, 28 Jun 1995 22:22:57 +1000 (EST)
From: Matthew Lamari <Matthew DOT Lamari AT jcu DOT edu DOT au>
To: djgpp AT sun DOT soe DOT clarkson DOT edu
Subject: Very Nasty V2 Residue Bug? (who knows)

    This one is a little hard to explain.  I have this program, bit of 3d 
garb etc., that I had running fine in dpmi, under windows, whatever, 
under v1.1x etc.  A simple test I had for the routines was loading an 
object and spinning it across the 3D view from left to right, with a for 
loop that runs from -300 to 300 (x value)

   Thing is, on the occasions where i can get this to run, it moves the 
object across left to right, starting at the left.  But when I break from 
the program and rerun it, the program continues from where it left off 
last time.  Of course, it usually crashes after using the compiler, 
before running; but a clean boot lets it run.  Could that be some residue 
from the compiler running that blocks it out?

   When I slap it on the debugger it dies at different sections, for no 
apparent reason.  I'm sure the code is basically sound (i.e. no 
segmentation faults possible where it crashes) but it just does.  I'll 
state again, this program was fully dpmi compliant and the 3D engine was 
used thoroughly in another purpose, without failure.

   Could I be missing something obvious, or is this a nasty little bug?  
I understand that the compilers all worked fine.  I'd really like to get 
this resolved, so that I can get right into the true benefits of V2 (BTW 
all concerned - I think it's the best!)

   BTW, this effect was also noticeable when I booted without memory 
management at all (except for a VERY curious HEAVY (10 times slower) 
speed lag.

Keep up the good work, DJ etc.
   Hope you can find time to help a dummy! :)

Matthew.