From owner-freebsd-chat  Sun Aug 10 10:42:41 1997
Return-Path: <owner-freebsd-chat>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id KAA02173
          for chat-outgoing; Sun, 10 Aug 1997 10:42:41 -0700 (PDT)
Received: from kithrup.com (kithrup.com [205.179.156.40])
          by hub.freebsd.org (8.8.5/8.8.5) with SMTP id KAA02168
          for <chat@freebsd.org>; Sun, 10 Aug 1997 10:42:39 -0700 (PDT)
Received: (from sef@localhost) by kithrup.com (8.6.8/8.6.6) id KAA17395; Sun, 10 Aug 1997 10:42:38 -0700
Date: Sun, 10 Aug 1997 10:42:38 -0700
From: Sean Eric Fagan <sef@Kithrup.COM>
Message-Id: <199708101742.KAA17395@kithrup.com>
To: chat@freebsd.org
Subject: Re: variable sized arrays and gcc 
In-Reply-To: <11989.871200795.kithrup.freebsd.chat@time.cdrom.com>
References: Your message of "Sun, 10 Aug 1997 03:22:28 EDT."
             <199708100722.DAA03236@skynet.ctr.columbia.edu> 
Organization: Kithrup Enterprises, Ltd.
Sender: owner-freebsd-chat@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

In article <11989.871200795.kithrup.freebsd.chat@time.cdrom.com> you write:
>This has been a gcc feature for ages (and I mean years) but I've
>always avoided it in favor of an explicit alloca() call for the same
>reason - I don't want code which walks that far outside ANSI's painted
>lines.

alloca is not ANSI either, Jordan :).  Not all systems provide it; not all
compilers really have support for it.  (You can use Doug Gwyn's "portable"
one, but that's iffy.)

alloca() was the reason gcc added VLA's in the first place, actually.  And
the code is the same.

The scary part about VLA's is that they mean sizeof() is no longer a
evaluated-only-at-compile-time operator.