From owner-freebsd-hackers  Fri Jun 20 10:45:22 1997
Return-Path: <owner-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id KAA14689
          for hackers-outgoing; Fri, 20 Jun 1997 10:45:22 -0700 (PDT)
Received: from Sisyphos.MI.Uni-Koeln.DE (Sisyphos.MI.Uni-Koeln.DE [134.95.212.10])
          by hub.freebsd.org (8.8.5/8.8.5) with SMTP id KAA14679;
          Fri, 20 Jun 1997 10:45:07 -0700 (PDT)
Received: from x14.mi.uni-koeln.de (annexr2-38.slip.Uni-Koeln.DE) by Sisyphos.MI.Uni-Koeln.DE with SMTP id AA20916
  (5.67b/IDA-1.5); Fri, 20 Jun 1997 19:44:56 +0200
Received: (from se@localhost) by x14.mi.uni-koeln.de (8.8.5/8.6.9) id TAA21304; Fri, 20 Jun 1997 19:44:56 +0200 (CEST)
X-Face: "<d]#=8pzx);RzeqSKI86OVa7=!0/(uRa.+B.9Z9\eNUn@UG?!`y7yt2dFNn%k4'.}](uE%
 yCO)$e&Y1%3EO~ifu6Q-#YUM&JZ't,}JkPnAz,8Dj33u%@GBi%[Y#LHz$]h7a<p4)-jKI7~sKjlP-^
 EvA[G;]v&0]W!EL%shs,{7x0|oqN4YVIs5,NI#,V{9"WF):5&RkOhyj*#-IAG}Tnu;YCF,d
Message-Id: <19970620194455.34192@mi.uni-koeln.de>
Date: Fri, 20 Jun 1997 19:44:55 +0200
From: Stefan Esser <se@FreeBSD.ORG>
To: Troy Curtiss <troyc@sandy.merix.com>
Cc: freebsd-hackers@FreeBSD.ORG, Stefan Esser <se@FreeBSD.ORG>
Subject: Re: USB -- How tricky will it be to support?
References: <troyc-9705201538.AA000216302@sandy>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 0.74
In-Reply-To: <troyc-9705201538.AA000216302@sandy>; from Troy Curtiss on Fri, Jun 20, 1997 at 08:38:46AM -0800
Sender: owner-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

On Jun 20, Troy Curtiss <troyc@sandy.merix.com> wrote:
> Hackers,
>   How tricky is USB going to be to support?  Some of the mechanisms
> such as hot plug/unplug, auto-detection and driver allocation, addressing,
> and etc. don't look easy... The paradigm feels like a cross between the
> PC-Card stuff (for the plug/unplug) and a PCI-bus (for addressing and
> detection).

I have looked a little at USB, and found that the most tricky
part is support of isochronous devices (sound cards, modems, ...)

Probing and attaching an USB chip is not hard, but USB is more
like SCSI than a network driver: There are many classes of devices
which all use their own command sets. We'll need a well designed
"generic" layer, which has to take e.g. time synchronization of
USB devices into account.

In fact, there are quite detailed instructions on how a USB driver
should interface to such a generic layer. I don't like the style
of the code examples given in the USB UHCI (Universal Host Controller
Interface) specification, but at least that a start ... :)

> Anybody toyed with supporting USB yet?  I'm curious what angle to
> attack this subsystem from.

Well, start by reading all available USB specs. Then take a look 
at some implementation of a USB bridge (i.e. Intel PIIX-3 or -4).

Then tell us about your proposed architecture, best if you include
a working prototype :)

I'd be insterested to work on this, but have (too many) higher 
priority tasks, currently. But at least, I think we should start
a discussion on how to attack USB ...

Regards, STefan