From owner-freebsd-hackers  Mon Feb 26 22:23:12 1996
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id WAA01455
          for hackers-outgoing; Mon, 26 Feb 1996 22:23:12 -0800 (PST)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id WAA01370
          Mon, 26 Feb 1996 22:22:14 -0800 (PST)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.12/8.6.9) id RAA14236; Tue, 27 Feb 1996 17:17:44 +1100
Date: Tue, 27 Feb 1996 17:17:44 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199602270617.RAA14236@godzilla.zeta.org.au>
To: julian@ref.tfs.com, pst@cisco.com
Subject: Re: gratuitous changes to db/hash.c for threadsafe operation?
Cc: current@FreeBSD.org, hackers@FreeBSD.org, hsu@FreeBSD.org,
        julian@FreeBSD.org
Sender: owner-hackers@FreeBSD.org
Precedence: bulk

>It clashes with the errno in the thread_safe libc
>which is a MACRO
>#define errno (*__errno(current_thread))

>or something similar
>this is true in almost every threads package in the world...

ANSI permits errno to be a macro (to allow thngs like the above), so
it shouldn't be used in portable code to mean anything other than the
ANSI errno.  Similarly for almost all ANSI names, e.g., sin.

Bruce