Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 01 Aug 2002 16:44:46 -0400
From:      "Louis A. Mamakos" <louie@TransSys.COM>
To:        Josef Karthauser <joe@tao.org.uk>
Cc:        Dag-Erling Smorgrav <des@ofug.org>, cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG
Subject:   Re: cvs commit: src/sys/conf files src/sys/modules Makefile src/sys/modules/uvisor Makefile src/sys/dev/usb uvisor.c 
Message-ID:  <200208012044.g71KikkM078775@whizzo.transsys.com>
In-Reply-To: Your message of "Tue, 30 Jul 2002 23:18:36 BST." <20020730221836.GA12286@genius.tao.org.uk> 
References:  <200207301744.g6UHiTsh025084@freefall.freebsd.org> <xzpheihf4sz.fsf@flood.ping.uio.no> <200207302158.g6ULw70L015968@whizzo.transsys.com> <20020730221836.GA12286@genius.tao.org.uk> 

next in thread | previous in thread | raw e-mail | index | archive | help
> On Tue, Jul 30, 2002 at 05:58:07PM -0400, Louis A. Mamakos wrote:
> > > Josef Karthauser <joe@FreeBSD.org> writes:
> > > >   Log:
> > > >   Commit a version of the uvisor driver for connecting Handspring
> > > >   Visors via USB.
> > > 
> > > Shouldn't this rather be done in userland, through ugen?
> > 
> > In fact, it is available in userland code as part of code I 
> > contributed to the coldsync package (also in the FreeBSD ports
> > tree).
> 
> Coldsync doesn't work with my M505, at least the version in the
> ports doesn't.  The devel version in their cvs repository doesn't
> compile against FreeBSD (current).

I guess that my point was that trying to develop support for these
other platforms might be easier in user-mode code, but maybe that's
just me.

> > The protocol used with the Handspring Visor and Visor Deluxe
> > was reversed engineered with a USB protocol analyzer and is 
> > essentially the same as the serial line protocol but using
> > a USB pipe to carry the bits.
> 
> It also has some nice "undocumented" handshaking at the beginning
> that no-one appears to know what to do with.

At least for the Visor and Visor deluxe, the coldsync code that I
added uses that handshaking.  It's essentially a portmapper-like
mechanism to map functions (e.g., hotsync protocol) to USB endpoint
addresses.   The code in coldsync has some comments about this
based on some email I got out of the Handspring people.   I do not
know what's been changed in the later Handspring, Palm, Sony, etc.
Palm platforms.

> > Other USB-attached Palm-like devices do something different
> > which doesn't work with the code that I wrote.  Having this
> > in the kernel is unlikely to make it *easier* to support, but
> > what the heck.  The kernel driver will presumably also work with
> > the other legacy PalmOS sync packages that expect at TTY device.
> 
> Either way, it's there now.  Use it or choose not to ;).  Tools not
> policy.

Good luck to you all; I had my fill of kernel panics debugging the
early ugen driver in the kernel :-)

louie


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200208012044.g71KikkM078775>