Date: Wed, 25 Aug 1999 14:40:18 +0300 (IDT)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: DJ Delorie <dj AT delorie DOT com>
cc: djgpp-workers AT delorie DOT com
Subject: Re: djgpp 2.03 990819
In-Reply-To: <199908200228.WAA08090@envy.delorie.com>
Message-ID: <Pine.SUN.3.91.990825143805.2834h-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp-workers AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk


On Thu, 19 Aug 1999, DJ Delorie wrote:

> Is on its way.  I did a cvs get in a blank directory, it built on the
> first try, built zips in one try, and here they are!

This alpha has some problems:

 1) The CVS/* files are still there.  I send in a separate message the
    list of all files I believe should not be in the distribution
    (there are some additional files, beyond */CVS/*).

 2) kb.info was produced without --no-split, which generates files
    kb.info-1 and kb.info-2 (that aren't included in djdev).  I fixed
    the relevant makefile.  This was so in the previous alpha as well,
    I just missed it.  (Previously, kb.info was small enough, but now
    it is larger, and makeinfo splits it.)

 3) src/empty.s and src/libc/compat/math/exp{2,10}.s are still not in
    djlsr.

 4) It seems like the cross-assembler is still not patched, since the
    libraries are again bloated with local labels.

I notice that exp2.s and exp10.s are compiled as exp2.S and exp10.S,
accordingly (because makefile.inc defines the .S.o rule before .s.o).
This is harmless on DOS/Windows, and on Unix works as God intended, so
I guess we can leave this alone.