Replied: Tue, 21 Jul 1998 15:51:01 EDT
Replied: "Christoph Badura <bad@ora.de> "
Received: from SOUTH-STATION-ANNEX.MIT.EDU by po7.MIT.EDU (5.61/4.7) id AA25170; Tue, 21 Jul 98 14:53:07 EDT
Received: from orade.ora.de by MIT.EDU with SMTP
	id AA07328; Tue, 21 Jul 98 14:55:47 EDT
Received: (from bad@localhost)
	by orade.ora.de (8.8.8/8.8.8) id UAA29928;
	Tue, 21 Jul 1998 20:55:51 +0200 (MET DST)
Date: Tue, 21 Jul 1998 20:55:51 +0200 (MET DST)
From: Christoph Badura <bad@ora.de>
Message-Id: <199807211855.UAA29928@orade.ora.de>
To: nathanw@MIT.EDU
Subject: Re: Survey: PCMCIA modems and other com devices
Newsgroups: hanse-ml.netbsd.current-users
References: <199807160521.BAA22175@road-runner.mit.edu>

>	Our PCMCIA code is a little weird with respect to serial
>devices. The PCMCIA specification, as far as I know, does not provide
>a way to indicate a generic property like "com-port-compatible". So
>our driver for such devices depends on (1) a table of known serial devices
>and (2) a heuristic for identifying other devices. 

Uh, the function identification tuple clearly distinguishes serial ports.
And from a function extension tuple you can event get the type of
UART on the card.

All modulo vendor bugs, of course.

-- 
Christoph Badura
Verlag O'Reilly
