Date: Mon, 20 Sep 2004 14:27:33 +0300 From: Valentin Nechayev <netch@lucky.net> To: Poul-Henning Kamp <phk@phk.freebsd.dk> Cc: arch@freebsd.org Subject: Re: [HEADSUP] naming of tty devices. Message-ID: <20040920112733.GE84228@lucky.net> In-Reply-To: <46041.1095665925@critter.freebsd.dk> References: <46041.1095665925@critter.freebsd.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
Mon, Sep 20, 2004 at 09:38:45, phk wrote about "[HEADSUP] naming of tty devices.":
> My suggestion is the following:
> All drivers will offer "tty${something}" devices, and
> generally ${something} will consist of a letter followed
> by a number, possibly in base 36 ([0-9a-z]).
There are cases when cua devices are used as login devices, and too many
places where 3 first chars of device name are removed without checking them,
e.g. /bin/ps:
=== cut ===
else {
if (strncmp(ttname, "tty", 3) == 0 ||
strncmp(ttname, "cua", 3) == 0)
ttname += 3;
=== end cut ===
When restoring terminal device from this name, now it is possible yet
to determine terminal testing /dev/tty$x or /dev/cua$x; with your new
scheme this will be impossible totally.
If we suppose to remain ability of logging in via callout devices,
these code pieces must be rewritten to keep "tty"/"cua", or all callout
devices must be renamed to terminals.
On 4.* I use the latter variant on one host: /dev/cuaa$n are renamed
to /dev/ttyOd$n, n=0...3.
> This results in the following major compatibility issues:
> sio's cuaa* gets renamed to cuad*
Basing on written above I propose /dev/ttyOd*
-netch-
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040920112733.GE84228>