From: sandmann AT clio DOT rice DOT edu (Charles Sandmann)
Message-Id: <10108041846.AA14518@clio.rice.edu>
Subject: Re: Bash 2.05 buffer overrun problem
To: eliz AT is DOT elta DOT co DOT il
Date: Sat, 4 Aug 2001 13:46:24 -0500 (CDT)
Cc: pjfarley AT dorsai DOT org, djgpp-workers AT delorie DOT com, snowball3 AT bigfoot DOT com,
        acottrel AT ihug DOT com DOT au
In-Reply-To: <8296-Sat04Aug2001211344+0300-eliz@is.elta.co.il> from "Eli Zaretskii" at Aug 04, 2001 09:13:45 PM
X-Mailer: ELM [version 2.5 PL2]
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

> > From: sandmann AT clio DOT rice DOT edu (Charles Sandmann)
> > Date: Sat, 4 Aug 2001 12:58:36 -0500 (CDT)
> > 
> > If you want to be safe, dynamically allocate or make the size 32Kb.
> 
> Do we have any reason to believe that 32KB is the limit?

I suppose it could be 64Kb if they used the entire word for length ?

I did an experiment - created 36 environment variables of 2040 characters
each.  Yes, Windows 2000 is happy with that.  When I ran a DJGPP image
all of those environment variables were not there - it seems to have only
used 15Kb of environment space.

But it's clear NT isn't limited to 32K (or even 64K) of total environment
space and you don't have to pre-allocate the size.

I haven't figured out a clever way to test the maximum environment variable
size yet.

I'd say malloc it and don't worry about it.