X-Authentication-Warning: delorie.com: mailnull set sender to djgpp-bounces using -f
Message-ID: <3C6D62E3.34091E03@yahoo.com>
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
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Malloc/free DJGPP code
References: <Pine DOT SUN DOT 3 DOT 91 DOT 1020214175634 DOT 1382J-100000 AT is> <3C6C735D DOT A6D16719 AT yahoo DOT com> <1225-Fri15Feb2002084500+0200-eliz AT is DOT elta DOT co DOT il> <3C6CC93C DOT 6F9E05E6 AT yahoo DOT com> <2947-Fri15Feb2002112947+0200-eliz AT is DOT elta DOT co DOT il> <3C6CF755 DOT CC35209E AT yahoo DOT com> <7680-Fri15Feb2002142924+0200-eliz AT is DOT elta DOT co DOT il>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Lines: 35
Date: Fri, 15 Feb 2002 19:36:02 GMT
NNTP-Posting-Host: 12.90.173.123
X-Complaints-To: abuse AT worldnet DOT att DOT net
X-Trace: bgtnsc05-news.ops.worldnet.att.net 1013801762 12.90.173.123 (Fri, 15 Feb 2002 19:36:02 GMT)
NNTP-Posting-Date: Fri, 15 Feb 2002 19:36:02 GMT
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

Eli Zaretskii wrote:
> 
> > From: CBFalconer <cbfalconer AT yahoo DOT com>
> > Newsgroups: comp.os.msdos.djgpp
> > Date: Fri, 15 Feb 2002 11:58:27 GMT
> >
> > But that doesn't explain why the static 'firstime' control failed?
> 
> I think it failed because malloc is called from the startup code more
> than once.  So the condition guarded the very first call, but not the
> others.
> 
> > The file system to receive 'write' has no more reason to be
> > initialized at that point than does cprintf, does it?
> 
> The call to `write' doesn't need any initializations, assuming you
> write to handles 1 or 2, which are already preconnected when the
> program starts.  `write' is more or less straightforward path to a DOS
> call (you can use `_write' if you want to be even closer to the DOS
> call, but in that case, you will need to use "\r\n" to end a line).

Finally I have debug output.  The combination of sprintf to a
local buffer and write actually works.  You were right, there are
a lot more calls to sbrk during initialization (around 40), but I
guess the clobbered output mechanisms could never show them, so I
thought there was just the one.

This way output *CAN* be redirected.

-- 
Chuck F (cbfalconer AT yahoo DOT com) (cbfalconer AT XXXXworldnet DOT att DOT net)
   Available for consulting/temporary embedded and systems.
   (Remove "XXXX" from reply address. yahoo works unmodified)
   mailto:uce AT ftc DOT gov  (for spambots to harvest)