From owner-cvs-lib  Thu Dec  5 11:11:26 1996
Return-Path: <owner-cvs-lib>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.4/8.8.4) id LAA25077
          for cvs-lib-outgoing; Thu, 5 Dec 1996 11:11:26 -0800 (PST)
Received: from time.cdrom.com (time.cdrom.com [204.216.27.226])
          by freefall.freebsd.org (8.8.4/8.8.4) with ESMTP id LAA25067;
          Thu, 5 Dec 1996 11:11:20 -0800 (PST)
Received: from time.cdrom.com (localhost [127.0.0.1]) by time.cdrom.com (8.8.4/8.6.9) with ESMTP id LAA12451; Thu, 5 Dec 1996 11:11:18 -0800 (PST)
To: Paul Traina <pst@shockwave.com>
cc: "Jordan K. Hubbard" <jkh@freefall.freebsd.org>,
        CVS-committers@freefall.freebsd.org, cvs-all@freefall.freebsd.org,
        cvs-lib@freefall.freebsd.org
Subject: Re: cvs commit: src/lib/libc/net getservent.c 
In-reply-to: Your message of "Thu, 05 Dec 1996 10:53:51 PST."
             <199612051853.KAA03962@precipice.shockwave.com> 
Date: Thu, 05 Dec 1996 11:11:18 -0800
Message-ID: <12447.849813078@time.cdrom.com>
From: "Jordan K. Hubbard" <jkh@time.cdrom.com>
Sender: owner-cvs-lib@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

> Since I'm being so contrarian today, -Wall and -pedantic will find 95% of the
> same stuff that ParaSoft is finding, and we have it today.

Actually, not in this case.  I did actually think of that, and ran
some comparative runs (my first question too: "I wonder how this
compares to -Wall").  -Wall finds lots of "might be uninitialized"
cases that Insure++ seems to ignore, but Insure++ in turn finds a lot
more bogus casts and gratuitious address-of usage.  It also caught the
long != int declarations where gcc was strangely content.

If the runtime also worked, I'd probably be finding a lot more
malloc/bounds error stuff, it just seems to have trouble with our
symbol tables vs BSD/OS's ones for now.

					Jordan