udevd broken on 3.5.4.1-alpha1 akita ?

Hi.

I've noticed that udevd doesn't modprobe the modules corresponding to
the devices inserted in the USB host...

Using udemonitor --env I see only UEVENT messages and no UDEV
ones... weird, according to how it is supposed to work
(http://vrfy.org/log/recent-state-of-udev.html)

I tried and watch more debug info, hence : 
udevcontrol log_priority=info

Then logread :
Mar 20 15:25:05 akita user.info kernel: usb 1-2.1: new full speed USB device using pxa27x-ohci and address 16
Mar 20 15:25:05 akita daemon.info udevd[3560]: udev_event_run: seq 872 forked, pid [4843], 'add' 'usb', 0 seconds old
Mar 20 15:25:05 akita daemon.info udevd-event[4843]: udev_event_run: seq 872 finished
Mar 20 15:25:05 akita daemon.info udevd[3560]: udev_done: seq 872, pid [4843] exit with 0, 0 seconds old
Mar 20 15:25:05 akita daemon.info udevd[3560]: udev_event_run: seq 873 forked, pid [4844], 'add' 'usb', 0 seconds old
Mar 20 15:25:05 akita daemon.info udevd[3560]: udev_event_run: seq 874 forked, pid [4845], 'add' 'usb', 0 seconds old
Mar 20 15:25:05 akita daemon.info udevd[3560]: udev_event_run: seq 875 forked, pid [4846], 'add' 'usb', 0 seconds old
Mar 20 15:25:05 akita daemon.info udevd-event[4844]: udev_event_run: seq 873 finished
Mar 20 15:25:05 akita daemon.info udevd[3560]: udev_done: seq 873, pid [4844] exit with 0, 0 seconds old
Mar 20 15:25:05 akita daemon.info udevd-event[4845]: udev_event_run: seq 874 finished
Mar 20 15:25:05 akita daemon.info udevd[3560]: udev_done: seq 874, pid [4845] exit with 0, 0 seconds old
Mar 20 15:25:05 akita daemon.info udevd-event[4846]: udev_event_run: seq 875 finished
Mar 20 15:25:05 akita daemon.info udevd[3560]: udev_done: seq 875, pid [4846] exit with 0, 0 seconds old
Mar 20 15:25:05 akita daemon.info udevd[3560]: udev_event_run: seq 876 forked, pid [4847], 'add' 'usb_device', 0 seconds old
Mar 20 15:25:05 akita daemon.info udevd-event[4847]: run_program: '/bin/sh -c 'X=usbdev1.16 X=${X#usbdev} B=${X%%.*} D=${X#*.}; echo bus/usb/$B/$D''
Mar 20 15:25:05 akita daemon.info udevd-event[4847]: run_program: '/bin/sh' (stdout) 'bus/usb//'
Mar 20 15:25:05 akita daemon.info udevd-event[4847]: run_program: '/bin/sh' returned with status 0
Mar 20 15:25:05 akita daemon.info udevd-event[4847]: udev_rules_get_name: add symlink 'bus/usb//'
Mar 20 15:25:05 akita daemon.info udevd-event[4847]: udev_rules_get_name: no node name set, will use kernel name 'usbdev1.16'
Mar 20 15:25:05 akita daemon.info udevd-event[4847]: create_node: creating device node '/dev/usbdev1.16', major = '189', minor = '15', mode = '0660', uid = '0', gid = '0'
Mar 20 15:25:05 akita daemon.info udevd-event[4847]: create_node: creating symlink '/dev/bus/usb//' to '../../../usbdev1.16'
Mar 20 15:25:05 akita daemon.err udevd-event[4847]: create_node: symlink(../../../usbdev1.16, /dev/bus/usb//) failed: File exists
Mar 20 15:25:05 akita daemon.info udevd-event[4847]: udev_event_run: seq 876 finished
Mar 20 15:25:05 akita daemon.info udevd[3560]: udev_done: seq 876, pid [4847] exit with 0, 0 seconds old

Actualy, there is indeed a problem in creating the symlink...

I suspect that this failing, it stops further processing, so no
modprobe will be done.

FYI,  udemonitor --env provides traces of valid MODALIAS values (for
instance :
MODALIAS=usb:v04BFp0319d0524dcE0dsc01dp01icE0isc01ip01), values which
work in modprove :
# modprobe -v usb:v04BFp0319d0524dcE0dsc01dp01icE0isc01ip01
insmod /lib/modules/2.6.15/kernel/net/bluetooth/bluetooth.ko
insmod /lib/modules/2.6.15/kernel/drivers/bluetooth/hci_usb.ko

I'll try and have a look at udev's source.

In the meantime, if someone has any clue...

Best regards,
-- 
Olivier BERGER (OpenPGP: 1024D/B4C5F37F)
APRIL (http://www.april.org) - Vive python (http://www.python.org)
Pétition contre les brevets logiciels : http://petition.eurolinux.org


-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid0944&bid$1720&dat1642