From owner-freebsd-current  Fri Feb 21 00:41:55 1997
Return-Path: <owner-current>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id AAA14131
          for current-outgoing; Fri, 21 Feb 1997 00:41:55 -0800 (PST)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA14126
          for <freebsd-current@freebsd.org>; Fri, 21 Feb 1997 00:41:51 -0800 (PST)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id TAA01669; Fri, 21 Feb 1997 19:35:03 +1100
Date: Fri, 21 Feb 1997 19:35:03 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199702210835.TAA01669@godzilla.zeta.org.au>
To: bde@zeta.org.au, jmg@nike.efn.org
Subject: Re: bzero and FD_ZERO
Cc: freebsd-current@freebsd.org
Sender: owner-current@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

>hmm...  I did a quick nm of libc.a, looks like the only bzero call is
>normal bzero...  same with memset...  is it time that we possible break
>these functions into a "specialized" header?  something like memfcn.h that
>will hold b* and mem* set of functions?

Lite2 changed a lot of string function calls to ANSI versions.  Wait 
until we catch up.  Putting the BSD versions in a separate header
wouldn't help much.  We would become incompatible with BSD.

>it looks like both sys/types.h and sys/buf.h require bzero...
>machine/pcaudioio.h requires memset... and machine/soundcard.h on
>memcpy...

sys/types.h is the only Standard header here.  sys/buf.h is a kernel-only
header.  The others probably should be documented as requiring <string.h>
if certain ioctls are used (if they are documented at all :-).

Bruce