OpenEuphoria

Re: EU 3.0.1 -- there's probably a bug in there somewhere

• Posted by Robert Craig <rds at RapidEuphoria.com> Feb 01, 2007
• 782 views

Andy Serpa wrote:
> Some of my code that runs fine on Eu 2.5 crashes with a machine exception with
> Eu 3.0.1.  I've verified that I'm not doing anything I shouldn't in terms of
> allocating/poking/freeing, etc, so I don't believe I'm corrupting any memory.
>  It always occurs when executing the return statement of a function, so I
>  imagine
> there is a reference-counting problem somewhere or some such thing that is
> over
> my head.
> 
> The programs this occurs with are way too complex to try to demonstrate the
> problem, and so I don't have a repeatable simple example I can show (although
> these errors are repeatable and consistent), but they do involve manipulating
> large sequences and complicated operations.
> 
> I did try running the code through the Euphoria-coded interpreter (exw.exe ->
> eu.exw -> myprogram.exw) and it did not crash.  Also translated versions of
> the same code run fine as well.
> 
> So anyone who knows what they are looking at might want to examine whatever
> the interpreter does when exiting a function that is different from v2.5 and
> see if they can find anything.  This is 2.5 compatible code -- no tasking
> stuff
> in there at all.  If I'm able to come up with a simple example I'll post it...

Yes, I'd like to see it, if you can set up a simple example.

So far no one has reported any similar crashes to yours.
If you or anyone else is serious about tracking this down right away,
you might build a debug version of the exw.exe interpreter,
or at least one with some added print statements,
and try to find where exactly in the interpreter it is crashing.
At a return, there is often some storage recycling which could
crash if something on the heap has been corrupted (either by the
user program or the interpreter).

Regards,
   Rob Craig
   Rapid Deployment Software
   http://www.RapidEuphoria.com