FC5 + Palm TX: Haaaaalp!

[Pete Toscano]

Hello,

I installed FC5 soon after it came out and keep things updated on a
daily basis.  Not long after, I got things playing nicely with my Palm
T3.  For me, "playing nicely" meant that I could manipulate the T3 with
the pilot-link tools, specifically pilot-xfer.

Not long after that, I got a TX.  The first one was wonky, so I returned
it and got a second that seems to sync with Windows and OS X (with
missing sync) just fine, but I cannot for the life of me get it working
with FC5.

The first strangeness that I noticed was that when I plugged in the hot
sync cable, it immediately brings up /dev/ttyUSB0 and /dev/ttyUSB1.
None of my previous PalmOS devices did this.


kernel: usb 4-2: new full speed USB device using uhci_hcd and address 19
kernel: usb 4-2: configuration #1 chosen from 1 choice
kernel: drivers/usb/serial/usb-serial.c: USB Serial support registered
for Handspring Visor / Palm OS
kernel: drivers/usb/serial/usb-serial.c: USB Serial support registered
for Sony Clie 3.5
kernel: drivers/usb/serial/usb-serial.c: USB Serial support registered
for Sony Clie 5.0
kernel: visor 4-2:1.0: Handspring Visor / Palm OS converter detected
kernel: usb 4-2: Handspring Visor / Palm OS converter now attached to
ttyUSB0
kernel: usb 4-2: Handspring Visor / Palm OS converter now attached to
ttyUSB1
kernel: usbcore: registered new driver visor
kernel: drivers/usb/serial/visor.c: USB HandSpring Visor / Palm OS driver


Udev is configured to create the /dev/pilot link and links it to the
correct (at least for the T3) device:

lrwxrwxrwx 1 root root      7 Apr 11 11:49 /dev/pilot -> ttyUSB1
crw-rw---- 1 root uucp 188, 0 Apr 11 11:49 /dev/ttyUSB0
crw------- 1 pete root 188, 1 Apr 11 11:49 /dev/ttyUSB1

Of course, if I try "pilot-xfer -l" right now, it just hangs, which is
expected.

Now, when I hit the hotsync button, this is what I see in /var/log/messages:

kernel: usb 4-2: USB disconnect, address 19
kernel: visor ttyUSB0: Handspring Visor / Palm OS converter now
disconnected from ttyUSB0
kernel: visor ttyUSB1: Handspring Visor / Palm OS converter now
disconnected from ttyUSB1
kernel: visor 4-2:1.0: device disconnected
kernel: usb 4-2: new full speed USB device using uhci_hcd and address 20
kernel: usb 4-2: configuration #1 chosen from 1 choice
kernel: visor 4-2:1.0: Handspring Visor / Palm OS converter detected
kernel: usb 4-2: Handspring Visor / Palm OS converter now attached to
ttyUSB0
kernel: usb 4-2: Handspring Visor / Palm OS converter now attached to
ttyUSB1


It looks like the original /dev/ttyUSB[01] are disconnected and then
re-created.

A "pilot-xfer -l" to /dev/pilot and /dev/ttyUSB1 both lead the the same
hang I saw earlier, before the hotsync button on the cable was pressed.
 If I try to "pilot-xfer -l" to /dev/ttyUSB0, I get a
permissions-related smack-down, complete with a libc backtrace and
memory map.

I had read online about setting the vendor and product values in the
visor module's options in /etc/modprobe.conf ("options visor
vendor=0x830 product=0x61"), but that doesn't change much.  (The vendor
and product values are mentioned in the syslog messages when the module
is inserted.)

I had also read that it's now possible to sync using libusb ("-p usb:")
if the RPM is re-compiled with the right flag.  I tried this, but didn't
notice any change in behavior.  It sounds like an intriguing option, but
I just want normal sync to work before I worry about getting fancy.

Finally, I have read online (from David Desrosiers) that normal USB
syncing is working fine between Linux and the TX, so I'm guessing that
it's some brain damage on my behalf.  I could probably get WiFi or
Bluetooth sync working too, but since I went through the trouble of
getting a second cable and it's also (probably) the fastest method, I'd
rather get this working first.

Would somebody help me?  I'm tired of not being able to do a normal sync
in Linux.

Thanks,
pete