This is the mail archive of the
guile@sources.redhat.com
mailing list for the Guile project.
Re: GUILE's GC - why we struggling to solve already solved problems ? (was Re: finding the top of the stack in linux)
- To: Marius Vollmer <mvo at zagadka dot ping dot de>
- Subject: Re: GUILE's GC - why we struggling to solve already solved problems ? (was Re: finding the top of the stack in linux)
- From: Tom Tromey <tromey at cygnus dot com>
- Date: Mon, 17 Jul 2000 15:20:01 -0700 (PDT)
- Cc: "Khimenko Victor" <guile at khim dot sch57 dot msk dot ru>, hanwen at cs dot uu dot nl, guile at sourceware dot cygnus dot com, crayc at kiwi dot pyro dot net
- References: <14707.2884.864156.799999@dokkum.cs.uu.nl><ABlUpSv8P1@khim.sch57.msk.ru><878zv0scug.fsf@zagadka.ping.de>
Marius> While Guile's GC is technically a conservative GC because it
Marius> scans certain memory areas for possible pointers, it is in
Marius> fact mostly precise. Do you think the Boehm GC can be as
Marius> efficient even when it is completely conservative?
You can use the Boehm GC in a partially precise mode. That is what we
do for libgcj.
Marius> In general, my view is that the GC of Guile is in fact an area
Marius> where significant improvements can be made, but I see these
Marius> improvements mostly in the direction of using more `modern'
Marius> variants of GC than mark/sweep, like generational collectors,
Marius> and mostly copying ones. I would regard switching to a purely
Marius> conservative GC as a step in the wrong direction.
You aren't describing the Boehm GC here. It also has generational
capability. Boehm is working on making it concurrent as well.
Marius> In any case, we could probably raid it for useful portability
Marius> code, like code for finding the stack boundaries, if it
Marius> contains something like that.
It does have this. Be prepared to track bug fixes if you go this
route, though.
Tom