Crashing in maxima 5.13.0 on Debian amd64



dlakelan wrote:
> dlakelan wrote:
>> I am playing around with my simulated annealing code and getting a 
>> highly reproducible crash on my machine when I execute the following 
>> code. 
...

> Based on the responses both on and off list, I'm guessing that this code 
> somehow activates a bug in either the GCL garbage collector, or the 
> interpreter. I am downloading the 5.14.0 source and going to compile for 
> SBCL rather than continue to use the Debian package. 

I have installed 5.14.0 on SBCL on my machine, and I also converted the 
argument to exp into a float. The combination has fixed several problems.

1) My SA code now runs quickly and cleanly on all test optimizations 
that I have thrown at it (including some fairly complicated ones 
involving minimax type criteria).

2) C-c C-c in emacs now stops maxima without causing a crash. previously 
under GCL it would often complain about "memory may be corrupted" and 
indeed it would rarely proceed.

3) Things run faster in general it seems.

My vote is to move to SBCL as the standard platform as soon as a 32 bit 
windows version can be reliably had.