gcl-devel
[Top][All Lists]
Advanced

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

[Gcl-devel] Re: release


From: Camm Maguire
Subject: [Gcl-devel] Re: release
Date: 23 Jan 2002 18:33:40 -0500

address@hidden writes:

> I will try to build maxima with the mingw build, then I will let
> you know.  If it works I will put the build on my web site, and
> I will stop trying to get cygwin working.  As far as I am concerned,
> one version that will run on windows is ok with me.

Agreed.

> Also, why dont we ask Dr. Bruno Haible to take a look at gcl's garbage
> collection?  He is probably very familiar with it as he got it working
> for clisp in many different environments.

He would be a great reference.  Do you have any contact information,
or indication that he might have the time to advise?


> Lastly, here is a link someone emailed me on gc
> http://www.daimi.au.dk/~beta/Papers/Train/train.html

Interesting link, but its a bit far down the pipeline to think about
implementing.  As far as the gc goes, from what I can tell, we have
the following advantages:

1) exact, the hard work about identifying all possible pointer
   locations has been done.
2) very space efficient -- lisp objects of the same size are grouped
   onto the same page.  
3) Portable, in principle, given stack locations, etc.  All in
   cimpleable C code.

and disadvantages:

1) stop the world, -- not incremental
2) fragments memory space into different types, making it hard to manage
3) Performance (?) (unsure here, really)

Take care,

> Dan
> 
> 

-- 
Camm Maguire                                            address@hidden
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah



reply via email to

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