Date: Mon, 11 May 1998 10:01:42 +0200 (WET)
From: Andris Pavenis <pavenis AT lanet DOT lv>
To: Nate Eldredge <nate AT cartsys DOT com>
cc: Alexander Bokovoy <bokovoy AT bspu DOT ac DOT by>, djgpp AT delorie DOT com
Subject: Re: Re[2]: SWORD 2.50; Anybody else use it or thinking about it?
In-Reply-To: <19980509172947.AAI16984@ppp105.cartsys.com>
Message-ID: <Pine.A32.3.91.980511095407.82616B-100000@ieva01.lanet.lv>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Precedence: bulk



On Sat, 9 May 1998, Nate Eldredge wrote:

> At 09:52  5/8/1998 +0200, Alexander Bokovoy wrote:
> >Nate Eldredge wrote:
> >> At 09:30  5/7/1998 +0200, Alexander Bokovoy wrote:
> >>>I compiled February's edition of SWORD 3.0 under DJGPP with some
> >>>hacks. Main problem is broken exceptions support in GCC 2.8.0 port so
> >>>most of SWORD 3.0 tests doesn't work under DJGPP because of
> >>>intensive usage of exceptions. We looking for corrections to GCC's
> >>>port (Robert Hoehne is working under it but ... :-( )
> >> I was under the impression that exceptions worked fine, although less
> >> efficiently, if you compiled with `-fsjlj-exceptions'. Is that not so?
> >I don't test this yet but there was port of GCC 2.8.1 announced with
> >'right'-worked exceptions, so it seems like problem has been solved.
> >Anyway, thanks in advice.
> 
> As far as I could tell from reading the announcement, it just looked like
> `-fsjlj-exceptions' had been enabled by default in the `specs' file.
> 

I have worked with gcc-2.8.1 for DJGPP already for more than a month. The 
exceptions worked for me unless some errors were done while installing
gcc (I mentioned these possible errors in README.DJGPP). There were NO 
NEED to use -fsljl-exceptions as the correct way suggested at first in
libgcc.c and later by Robert were used. 

Andris