[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Axiom-developer] Sets in MonoidRing

From: root
Subject: Re: [Axiom-developer] Sets in MonoidRing
Date: Wed, 23 Jul 2003 15:57:30 -0400


In fact, the parallel trace idea is how I've tried to debug the
compiler problem. Unfortunately it is not really that simple
(there is no such thing as a simple job) as the code differs
in the two lisps. However for this case we can compare GCL vs
CMUCL which would be running exactly the same code.

You can trace anything you can get your hands on in any common 
lisp. The trick is to figure out where Axiom hides stuff. The
hardest part is that even at 2.5Ghz a computation can take a
second or so. That's 2.5 BILLION (U.S. :-) ) instructions that
can be wrong. I've had a parallel trace walk going for 3 days
chasing the compiler bug and still not cornered it. The system
is very complex and takes a lot of skill and patience to debug
even the simple things, mostly because the person who's job it
was (me) didn't document the damn thing. We're gonna fix that.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]