Email list hosting service & mailing list manager

(Previous discussion continued)
Re: Comparing Pika-syle and JNI-style Shiro Kawai (15 Jan 2004 00:30 UTC)

Re: Comparing Pika-syle and JNI-style Shiro Kawai 15 Jan 2004 00:30 UTC

>From: bear <xxxxxx@sonic.net>
Subject: Re: Comparing Pika-syle and JNI-style
Date: Wed, 14 Jan 2004 15:55:13 -0800 (PST)

> I've been bitten by this while using a genetic algorithm.  I had
> "random-looking" bit patterns (the genomes) filling almost 2Gbytes of
> memory space, and the Boehm collector was mistaking almost half of
> the words for possible pointers.  It eventually reached a tipping
> point where the values it couldn't free contained enough
> "pseudopointers" to prevent it from freeing anything else, and
> crashed.

I don't object to the remarks about the drawback of conservative GC;
but just curious about your specific experience.  Was there a
reason that prevented you from using GC_malloc_atomic?
(I usually carefully avoid allocating big chunk of non-pointer
data via GC_malloc, but if there's a general case that I can't
avoid it, I want to know that).

--shiro