(is this a kernel, doco, anaconda, or ... bug?) The DBT-120 adaptor doesn't work out-of-the-box. The driver reports; bcm203x_probe: Mini driver request failed bcm203x: probe of 1-1:1.0 failed with error -5 You need to fetch the firmware from: http://www.bluez.org/download.html and install the files: BCM2033-MD.hex BCM2033-FW.bin in /lib/firmware (the firmware tar ball will do it if you do: ./configure; make; make install).
FWIW, tried it on Rawhide/x86 (its almost FC4test2 now) Bluetooth: Core ver 2.7 NET: Registered protocol family 31 Bluetooth: HCI device and connection manager initialized Bluetooth: HCI socket layer initialized Bluetooth: HCI USB driver ver 2.8 usbcore: registered new driver hci_usb And it worked. I recall this working with Rawhide (pre-fc4t1) on powerpc as well, however I haven't rolled an install out in a while (and am away from my ppc boxes now) lsusb reports the adapter (a dbt-120) as: Bus 003 Device 003: ID 0a12:0001 Cambridge Silicon Radio, Ltd Bluetooth Dongle (HCI mode) IIRC, on irc, you reported your lsusb output as being: <cagney> Bus 001 Device 002: ID 0a5c:2033 Broadcom Corp. BCM2033 Bluetooth So guess this isn't the first time they release the same product model and change the internals of it with a different manufacturer..
PS: The above, while correct, isn't working. (Following DWMW2's pointer) I'm fi that /etc/hotplug/firmware.agent fails with: + cp /lib/firmware/BCM2033-MD.hex /sys//class/firmware/1-1/data cp: cannot create regular file `/sys//class/firmware/1-1/data': No such file or directory + echo 0 /etc/hotplug/firmware.agent: line 49: /sys//class/firmware/1-1/loading: No such file or directory
cf. http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=302990 http://marc.theaimsgroup.com/?l=linux-hotplug-devel&m=111087517321591&w=2 and potential patch in http://marc.theaimsgroup.com/?l=linux-hotplug-devel&m=111090287820444&w=2
could you try: ftp://people.redhat.com/harald/udev/udev-056-1.src.rpm
needs udev-056 and 2.6.12 or patch for 2.6.11 http://kernel.org/pub/linux/kernel/people/gregkh/gregkh-2.6/patches/driver/firmware-timeout.patch
I've also been having issues with firmware loading (ipw2100) and it turned out to be a problem with the initial event timeout in udevd: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=145283#c10 This too is fixed in udev-056, but doesn't require the kernel patch. So if firmware loading only fails during boot, it's likely a startup timeout issue. The kernel patch avoids the issue completely by allowing firmware events to bypass the udevd queue.
Harold, Should 2.6.11-1.1236_FC4 and udev-050-10 work (it doesn't appear to) or do I also need to also build / install the udev-056 rpm? $ uname -a Linux localhost.localdomain 2.6.11-1.1236_FC4 #1 Mon Apr 11 12:23:42 EDT 2005 ppc ppc ppc GNU/Linux $ rpm -q udev udev-050-10 $ sudo hciconfig $ dmesg | tail ADB keyboard at 2, handler 1 ADB mouse at 3, handler set to 4 (trackpad) adb: finished probe task... usb 1-1: new full speed USB device using ohci_hcd and address 2 Bluetooth: Broadcom Blutonium firmware driver ver 1.0 Bluetooth: HCI USB driver ver 2.8 bcm203x_probe: Mini driver request failed bcm203x: probe of 1-1:1.0 failed with error -5 usbcore: registered new driver bcm203x usbcore: registered new driver hci_usb $
Andrew, my name is Harald and please test udev-056.
Ok, I've tested udev-056-1 and finally get: # hciconfig hci0: Type: USB BD Address: 00:80:C8:50:F9:5C ACL MTU: 377:10 SCO MTU: 16:0 UP RUNNING PSCAN ISCAN RX bytes:101 acl:0 sco:0 events:13 errors:0 TX bytes:300 acl:0 sco:0 commands:13 errors:0 my phone sees my mac; but not the reverse; try again, ah there it is now for a dial out ...