From owner-freebsd-hardware  Sun Dec  1 20:49:26 1996
Return-Path: owner-hardware
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id UAA00249
          for hardware-outgoing; Sun, 1 Dec 1996 20:49:26 -0800 (PST)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id UAA00243
          for <hardware@FreeBSD.org>; Sun, 1 Dec 1996 20:49:22 -0800 (PST)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.3/8.6.9) id PAA10899; Mon, 2 Dec 1996 15:32:48 +1100
Date: Mon, 2 Dec 1996 15:32:48 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199612020432.PAA10899@godzilla.zeta.org.au>
To: bde@zeta.org.au, gerg@stallion.oz.au
Subject: Re: FreeBSD as Terminal Server
Cc: alex@yahoo.com, hardware@FreeBSD.org, sos@ravenock.cybercity.dk,
        sysseh@devetir.qld.gov.au
Sender: owner-hardware@FreeBSD.org
X-Loop: FreeBSD.org
Precedence: bulk

>> >>You'd need to translate the istallion driver from the Linlish :-).
>> ...

>Sure, but that one thing hardly makes the driver "Linlish".

See they smiley.

>> These drivers are mostly apparently unmaintained :=].
>
>Well, the Stallion drivers are???

Are they? :-)

I'd like all drivers to be actively maintained.  This seems to require
many commits per year even for "finished" drivers.  You could read the
commit logs for all changes to other tty drivers to make sure you're
not falling behind.

>I have modified the Stallion drivers to only use spltty, and am bench
>testing them now... Trivial change, code should be available in a few
>days.

Good.  Other things to do:

unfinished details:
- support devfs.
- support the drivers in MAKEDEV.
- convert README.stl into a man page.

recent changes to keep up with:
- support new ioctl TIOCDCDTIMESTAMP.  Better support old ioctl
  TIOCTIMESTAMP too.
- make sure that the drivers pass all NIST PCTS (POSIX) tests.  The
  tests have a good coverage of fancy termios stuff that no one uses :-).

future changes to keep up with:
- fix some ioctls that don't wait for the (controlling) tty to be in
  the foreground before changing it.  The wait is done in ttioctl(), but
  that is too late - it is after ttsetcompat(), locking and l_ioctl().
  This is most serious for ttsetcompat().  Locking inhibits unwanted
  changes and background processes are unlikely when the line discipline
  in nonstandard.  ttioctl() neglects to wait for many ioctls too.

Bruce