Message-Id: <3.0.6.32.19990202213636.00941100@pop.netaddress.com>
X-Sender: pderbysh AT pop DOT netaddress DOT com
X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.6 (32)
Date: Tue, 02 Feb 1999 21:36:36 -0500
To: djgpp AT delorie DOT com
From: Paul Derbyshire <pderbysh AT usa DOT net>
Subject: Re: Alignment fault signals
In-Reply-To: <199902022352.SAA14529@envy.delorie.com>
References: <3 DOT 0 DOT 6 DOT 32 DOT 19990202154617 DOT 0090ada0 AT pop DOT netaddress DOT com>
 <3 DOT 0 DOT 6 DOT 32 DOT 19990202120936 DOT 00906210 AT pop DOT netaddress DOT com>
 <3 DOT 0 DOT 6 DOT 32 DOT 19990202120936 DOT 00906210 AT pop DOT netaddress DOT com>
 <3 DOT 0 DOT 6 DOT 32 DOT 19990202154617 DOT 0090ada0 AT pop DOT netaddress DOT com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Reply-To: djgpp AT delorie DOT com

At 06:52 PM 2/2/99 -0500, you wrote:
>> Why not have an ANSI or at least POSIX signal macro for this...
>
>You'd have to ask the ANSI and/or POSIX committees about this.  I
>think it's just because they're so undefined anyway, there was no
>point putting them in the spec.  Why specify something that can't be
>specified?

That's not my intent -- rather, my intent is to specify a way to specify
something that might be specified. :-) That's what the ANSI signals and
POSIX extensions are for anyways -- to specify a bunch of signals related
to common circumstances, which a given system may or may not actually
generate but which are there for a portable program to check for.

-- 
   .*.  "Clouds are not spheres, mountains are not cones, coastlines are not
-()  <  circles, and bark is not smooth, nor does lightning travel in a
   `*'  straight line."    -------------------------------------------------
        -- B. Mandelbrot  |http://surf.to/pgd.net
_____________________ ____|________     Paul Derbyshire     pderbysh AT usa DOT net
Programmer & Humanist|ICQ: 10423848|