From owner-freebsd-hackers  Tue Aug 29 18:09:19 1995
Return-Path: hackers-owner
Received: (from majordom@localhost)
          by freefall.FreeBSD.org (8.6.11/8.6.6) id SAA01857
          for hackers-outgoing; Tue, 29 Aug 1995 18:09:19 -0700
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.34])
          by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id SAA01844
          for <hackers@freebsd.org>; Tue, 29 Aug 1995 18:08:53 -0700
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id LAA10577; Wed, 30 Aug 1995 11:07:16 +1000
Date: Wed, 30 Aug 1995 11:07:16 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199508300107.LAA10577@godzilla.zeta.org.au>
To: hackers@freebsd.org, peter@nmti.com
Subject: Re: Gritching about XFree86 and serial port naming
Sender: hackers-owner@freebsd.org
Precedence: bulk

>I'm helping a bloke here set up FreeBSD, and man the serial ports are hard
>to explain.

>Why, oh why, is the CU device for "ttyd0" called "cuaa0" instead of "cuad0"?
>Or making it ttya0?

It was to give a unique name when the name is truncated to 2 characters
for printing by ps, etc.  There isn't enough namespace for unique names,
so I wanted to change to cuad0, but the names have already changed too
oftenand further changes are likely and there wasn't much time before
2.0.5R.

>I understand that going to alpha names makes some sense, but it really doesn't
>look good when they're so inconsistent. The old names had the virtue of
>making sense "oh, sio is the serial I/O driver...".

When did they make sense?  They were never called sio* or ttys*.  `s'
would have been a bad name anyway since sio is actually the ns(x)xx50
driver.  This point becomes more important when there are half a dozen
other Serial IO drivers with different broken naming schemes.

>For 2.1, can we go back to at least making the tty and cua devices have the
>same identifiers?

When did they have the same identifiers?  If you say that tty00 and cua00
have the same identifiers, then I'll say that `0' is not an identifier :-).

>Oh yeh, the docs should have something in big bold letters about linking
>/dev/mouse to /dev/cuaa0 under whatever name it ends up as...

That would be bogus.  X works with the vanilla POSIX port ttyd0 and
always has.

Bruce