On Sun, 16 Jul 2006, Berndt Josef Wulf wrote:

> However, your patch seems to get us a step further see below:
> : {9} btconfig ubt0 up
> btconfig: SIOCSBTFLAGS: Resource temporarily unavailable
> ==========
> Jul 16 18:37:28  /netbsd: hci_usrreq: ATTACH
> Jul 16 18:37:28  /netbsd: hci_usrreq: CONTROL
> Jul 16 18:37:28  /netbsd: hci_ioctl: cmd 0xc0386205
> Jul 16 18:37:28  /netbsd: hci_usrreq: CONTROL
> Jul 16 18:37:28  /netbsd: hci_ioctl: cmd 0xc0386208
> Jul 16 18:37:28  /netbsd: ubt_enable: sc=0xc1f89400
> Jul 16 18:37:28  /netbsd: hci_send_cmd: (ubt0) opcode (004|0009)
> Jul 16 18:37:28  /netbsd: hci_send_cmd: (ubt0) opcode (004|0005)
> Jul 16 18:37:28  /netbsd: hci_send_cmd: (ubt0) opcode (004|0003)
> Jul 16 18:37:28  /netbsd: hci_event: (ubt0) event COMMAND COMPLETE
> Jul 16 18:37:28  /netbsd: hci_event_command_compl: (ubt0) opcode (004|0009)
> num_cmd_pkts = 1
> Jul 16 18:37:29  /netbsd: ubt_disable: sc=0xc1f89400
> Jul 16 18:37:29  /netbsd: ubt_abortdealloc: sc=0xc1f89400
> Jul 16 18:37:29  /netbsd: ubt_recv_acl_complete: status=CANCELLED (6)
> Jul 16 18:37:29  /netbsd: hci_usrreq: DETACH
>
> we are getting there....  :-)

Hm, but now that says to me something different, that the first command
only is succeeding and after that there is a blockage (when the three
commands are sent, the first will go to the device but the others will be
queued until the COMMAND_COMPLETE packet is received and num_cmd_pkts is
updated.)  I can't see any race condition in that, so I'm going back to
the thought that the problem is in the USB code..

can you increase the debug level and try the 'btconfig ubt0 up' again
please?

sysctl -w hw.ubt_debug=15
sysctl -w net.bluetooth.debug=10

should show where it is getting lost..

iain