Bluetooth headset GUI configuration

Carl Lobo carllobo at gmail.com
Sun Dec 21 14:39:10 CET 2008


Basically simply followed the steps on the Bluez Wiki for bonding. The
rest of Angus' scripts seem to work. You don't even need the device in
discoverable mode etc if you know it's bluetooth address
simply
passkey-agent --default <pin> (in one console - sometimes it stops on
tty output if i do an &)
dbus-send --system --type=method_call --print-reply --dest=org.bluez
/org/bluez/hci0 org.bluez.Adapter.CreateBonding string:<address>
then in ~/.asoundrc

pcm.headset_raw {
	type bluetooth
	device <address>
	profile "voice"
}

ctl.headset {
	type bluetooth;
	device <address>
}

pcm.headset {
	type plug
	slave.pcm = headset_raw
}

BtHeadset.py should work after that.

Now here's the strange thing - if I try

mplayer -ao alsa:device=headset it connects and shows me that its
playing - there are several card resets and hcid seg-faults every once
in a while, but I see SDP packets using hcidump. I tried all the Btcpu
state files as well as fiddling around with alsamixer (from whatever
little I could guess wildly at from the wolfson diagrams on wiki, past
emails etc) but no sound. There may be data coming in in a codec my
headset doesn't understand because if I press the volume key the
acknowledgement sound comes really distorted. Maybe my ScoRouting=HCI
then, I tried this quite some time back, I could have tried with
both...
however when I try a call -> the Angus' BtHeadset script shows me now
playing, same with hcid -n -d. However no sdp packets before or after
I restore the state. Also ScoRouting=PCM (otherwise it refuses the
headset play). And absolutely no sound on the call on either end (the
mic or speakers of the freerunner, headset or cordless phone). I can
hear both ends properly from the expected devices before I restore the
state file.
I tried building bluez 4.22 a couple of weeks back. It better
identifies the features of my headset (the full handsfree profile) but
doesn't look like it's implemented... (there's no telephony-*.c for
openmoko in the bluez source either). But no sound here as well and on
occasion a blinking red AUX led and a kernel crash :).
The headset does work with my laptop though - using bluez 3.30 I
think. I could hcidump on both devices if anyone can understand it and
help... At least the output of hcid -n -d seems very similar for both
the laptop (on FC9) and the FR (on the newest SHR).
The bluetooth headset is something I badly miss, I can't take any
calls while driving. But at least I can take calls pretty reliably now
though :)




More information about the community mailing list