Re: [coldsync-hackers] [FreeBSD / Clie] PEG-T415 behavior

["Cliff L. Biffle" <cbiffle at safety dot net>]

On Wednesday 06 March 2002 10:22 am, you wrote:
> I've tried with the S320, T415, and N610C and have had no problems
> whatsoever, under linux that is... sorry that doesn't help you out too much
> with freebsd.

Interesting!  So the difference probably lies in the fact that Linux has the 
visor kernel module, while FreeBSD does not (as far as I can tell).  I 
suspect that the visor mod sources would be quite useful here.

> I can say though that the driver used to support the Clie was also needed
> to support the M505+ and Handspring devices via USB, and have heard of some
> success using those under FreeBSD, using coldsync 2.2.5.

Really?  From what I can tell (from the dumps) the protocol used to 
communicate with the Clie (4.1) seems quite different from that used to 
communicate with my Visor Platinum (3.5.2).  I used to be able to sync my 
Plat against the BSD box here.  I wonder what the connection is?

> If you could dump here your .coldsyncrc, coldsync running while its
> printing debug output, as well as the log entries inside of
> /var/log/messages (or equivilant) that might help point out where it's
> failing.

I'll be happy to provide that data, but I'm not sure (in this case) how it's 
relevant...the failure occurs in the second third of the ritual packet 
exchange.  When using the default exchange in 2.2.5, the first transmission 
(from the Clie) is received properly, the second (to the Clie) is 
transmitted, but the Clie begins returning nasty invalid responses at that 
point (i.e. responding to a request 0x12 with a response 0x93 and other 
weirdnesses).  (Side note: I actually had to swap ritual_exch_server with 
ritual_exch_client in PConnection_usb.c to get the Clie to respond -at- -all- 
-- YMMV.)

With the ritual exchange commented out of the code, I was able to get more 
specific data on the exchange.  This is described in my other thread, but in 
short, the Clie starts off the exchange with an unrequested userinfo response 
packet, coldsync follows up with a sysinfo message, and the Clie then 
complains about the parameters.  The Clie appears to require that some 
specially-formatted  block of data be passed with the initial sysinfo 
request; from that point forward I was able to decode the entire exchange as 
standard DLP (albeit with some unexpected uses of long arguments).

In case it will help you, here's my coldsync.conf:
---
listen usb {
	protocol: net;
	device: /dev/ugen0;
}

pda "Clie" {
	snum: "H05205PV74KF-A";
	directory: /home/cbiffle/clie;
	username: "Cliff L. Biffle";
	userid: 2404;
}
---
I'm confident that the 'net' protocol is the right one, as it's the only one 
that has provoked -any- sort of response from the Clie.  The userid was 
retrieved from packet dumps of the exchange with Windows (it's returned, from 
what I can tell, as the second word-length field in a userinfo response 
[0x90]).  In any case, no data there is exchanged before the transaction 
errors out.

Any suggestions you might have are much appreciated!

-Cliff Biffle

-- 
This message was sent through the coldsync-hackers mailing list.  To remove
yourself from this mailing list, send a message to majordomo@thedotin.net
with the words "unsubscribe coldsync-hackers" in the message body.  For more
information on Coldsync, send mail to coldsync-hackers-owner@thedotin.net.