Message-ID: <3DB3CE9A.65E448AF@yahoo.com>
Date: Mon, 21 Oct 2002 05:53:30 -0400
From: CBFalconer <cbfalconer AT yahoo DOT com>
Organization: Ched Research
X-Mailer: Mozilla 4.75 [en] (Win98; U)
X-Accept-Language: en
MIME-Version: 1.0
To: djgpp-workers AT delorie DOT com
Subject: Re: CBFalconer's malloc
References: <200210210708 DOT JAA02518 AT lws256 DOT lu DOT erisoft DOT se>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Reply-To: djgpp-workers AT delorie DOT com

Martin Stromberg wrote:
> 
> > There is also the question of stress testing the code.  The easiest
> > thing to do would be to fix the bare minimum naming to be libc compatible,
> > dump it into a copy of libc.a, and try building the entire toolchain
> > twice to see if all goes well.  This would need a volunteer ...
> 
> I've built and rebuilt and rebuilt and rebuilt CVS libc and the tools
> (i. e. djgpp/src).

The makefile as it is will build a malloc.o, ready to be linked
into anything, using the names malloc, free, realloc.  I have
linked this ahead of the library to check operation during
initialization, with no problem.  I would expect that the only
addition needed is to build the library with that .o file.  IIRC
using such a configuration shows up the little stubs left from
initialization calls to malloc, which creates multiple unexpected
sbrk extensions. 

-- 
Chuck F (cbfalconer AT yahoo DOT com) (cbfalconer AT worldnet DOT att DOT net)
   Available for consulting/temporary embedded and systems.
   <http://cbfalconer.home.att.net>  USE worldnet address!