Message-Id: <200005092038.QAA10472@qnx.com>
Subject: Re: Perfomance of gc-simple
To: eliz AT is DOT elta DOT co DOT il
Date: Tue, 9 May 2000 16:37:37 -0400 (EDT)
From: "Alain Magloire" <alain AT qnx DOT com>
Cc: djgpp-workers AT delorie DOT com
In-Reply-To: <200005091906.PAA10862@indy.delorie.com> from "Eli Zaretskii" at May 09, 2000 03:06:15 PM
X-Mailer: ELM [version 2.5 PL0b1]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

> 
> > At the top of my head, I remember many applications on Un*x
> > use mmap() to grab some memory and maintain there own pool,
> 
> Yes, but why use mmap for that?  Why not use sbrk?

DJ answered that, and I beleive Solaris and GNU/Linux malloc
implementations are doing this. The QNX4 implemention does the
same two, it uses 2 allocators schemes base on the size of the
requested memory, one is call the band/arena so when you
release small chunks of memory it tries to collide them to make
a pagesize and call munmap() to give it back to the system.

On QNX/Neutrino sbrk() calls mmap()  ;-)

Theres probably other reasons.

> 
> > Was not there a VFS lib for DJGPP at one point ?
> 
> What's that?

http://www.geocities.com/CapeCanaveral/Lab/1862/djgpp/vfs_toc.html

I discover this while browsing the web.

-- 
au revoir, alain
----
Aussi haut que l'on soit assis, on n'est toujours assis que sur son cul !!!