From owner-freebsd-bugs  Sun Jul 23 23:23:13 1995
Return-Path: bugs-owner
Received: (from majordom@localhost)
          by freefall.cdrom.com (8.6.11/8.6.6) id XAA04347
          for bugs-outgoing; Sun, 23 Jul 1995 23:23:13 -0700
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.34])
          by freefall.cdrom.com (8.6.11/8.6.6) with ESMTP id XAA04340
          for <freebsd-bugs@freebsd.org>; Sun, 23 Jul 1995 23:23:05 -0700
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id QAA04367; Mon, 24 Jul 1995 16:18:30 +1000
Date: Mon, 24 Jul 1995 16:18:30 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199507240618.QAA04367@godzilla.zeta.org.au>
To: freebsd-bugs@freebsd.org, mpp@mpp.minn.net
Subject: Re: subyte()
Sender: bugs-owner@freebsd.org
Precedence: bulk

>I was looking into PR# 462 (uname system call doesn't return
>uts.version), and it appears to me that the problem is that
>the subyte() routine isn't writing the bytes back to the 
>correct address in the caller's memory.  Subyte() isn't
>called in very many places, so it is possible that
>it has been broken for a while and no one would have
>ever really noticed.

It is called a lot for tty input via ureadc().

>Could someone go double check me on this and make sure
>I'm not hallucinating about this?

The uname() in kern_xxx.c is a mirage.  It is now only for
compatibility.  uname() is now implemented in the library
by calling sysctl() for each component.

There is also compatibility cruft for a the BSDI uname().

Bruce