Date: Sat, 25 Aug 2001 14:58:07 +0300
From: "Eli Zaretskii" <eliz AT is DOT elta DOT co DOT il>
Sender: halo1 AT zahav DOT net DOT il
To: pavenis AT lanet DOT lv
Message-Id: <8011-Sat25Aug2001145806+0300-eliz@is.elta.co.il>
X-Mailer: Emacs 20.6 (via feedmail 8.3.emacs20_6 I) and Blat ver 1.8.9
CC: djgpp-workers AT delorie DOT com
In-reply-to: <3B87B702.13151.3BDE00@localhost> (pavenis@lanet.lv)
Subject: Re: gcc-3.0.1 WinXP and lfn=n
References: <10108241959 DOT AA13217 AT clio DOT rice DOT edu> (sandmann AT clio DOT rice DOT edu) <3B87B702 DOT 13151 DOT 3BDE00 AT localhost>
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: pavenis AT lanet DOT lv
> Date: Sat, 25 Aug 2001 14:32:34 +0300
> 
> Initially MD_EXEC_PREFIX is defined to /dev/env/DJDIR/bin. 
> Unfortunatelly it's canonicalized at startup (see definition in 
> gcc/config/i386/xm-djgpp.h). Perhaps I'll remove processing variable
> md_exec_prefix from this definition
> 
> So corresponding definition expands to current $DJDIR/bin at
> the time specs are dumped by xgcc ('gcc -dumpspecs').
> 
> Of course it would be possible to edit specs later by sed, but I would 
> like to avoid that

Sure.  If preventing the expansion of MD_EXEC_PREFIX solves this
problem, then I think it's an okay solution.  I don't think we need
/dev/env/DJDIR be expanded until the user runs the produced
executable.  There should be no need to expand it during the build or
dumpspecs stages.