Re: z22 not syncing anymore

[Tom Billiet <mouse256 ulyssis org>]

----- Original Message -----
From: Adam C Powell IV
Time: 07-09-07 21:57

Hello,

On Fri, 2007-09-07 at 21:08 +0200, Tom Billiet wrote:
  

Hi,

I have a problem that my palm z22 won't sync anymore with gnome-pilot.
It used to work, but for some reason an update broke it...
    

Update which OS/distro, from what version to what other?
  

I'm using archlinux, which has a "rolling release" system. That means that it has always the latest packages in the current repository. This also means that I can't answer to this question, I get continuous upgrades... It can be a kernel upgrade, it can be something else...

I've been trying to find the problem, but I really can't find it. I
also tried to switch from the visor kernel module (that's what I used
to use and worked) to libusb, but also no progress. I think the
problem is somewhere that the z22 makes 2 usb connections (USB0 and
USB1), and at the moment you hit the sync button he creates USB3 and
USB4 in favor of USB0 and USB1, gnome-pilot get's very confused of
this behavior.
So this is what I get with libusb. I start gpilotd, connect palm, give
it some seconds and hit the sync button (I always needed to give it a
couple of seconds, otherwise it wouldn't work)
    

This sounds vaguely like the behavior with my Treo 700p.  With this
device I cannot get visor to work at all.
  

Visor used to work, I have never been able to use libusb with the z22. Now none is working anymore.

With libusb, I need to control the timing very carefully: if I hit
hotsync on the Palm two to five seconds before the computer starts
listening, it works.  Less than two seconds before, or more than about
ten seconds before (never checked this timing precisely), or Palm after
computer, and it fails.  And by "fails" the gpilotd or pilot-xfer goes
into some kind of infinite loop as you described and needs to be killed.
So try a few different timings before/after and see if anything works.
  

Different timings do not give any better result, it stays the same...

It's easiest to check this by pausing the gpilotd daemon (kill it if
it's infinite-looping), and check the timing of something simple like
"pilot-xfer -p usb: -l".

Also, make sure it's not reloading the visor module, which would
interfere with libusb.

I have no idea what's actually happening which makes it fail/succeed...

Hope this helps,
-Adam
  

Thanks for your help, but so far no luck.

Tom