Hi,
i'm using bluez 3.36 on an embedded system with a CSR Chip connected.
Communication via HCI over UART works well. I need to receive and send
voice from/to a headset and therefore I want to send/get the voice as
PCM to the CSR chip. The PCM pins of the CSR Chip are connected to the
SSC pins of an avr32 microcontroller. The CSR chip is configured to use
PCM as soon as the Host produces Clock, otherwise it tries to transmit
voice per HCI which is too slow (only 115200 baudrate for a 8kHz PCM
stream).
In etc/bluetooth/audio.conf I have set SCORouting=PCM
In etc/Bluetooth/asound.conf I have written the following:
#####################
pcm.headset {
type bluetooth
}
ctl.headset {
type bluetooth
}
#####################
But bluez is still sending voice over HCI, which is ok because: How
should it know where to transmit the PCM stream. I need a device but I
am not sure what exactly is needed.
Do I have to write a dummy soundcard driver which only sends and
receives pcm over the ssc pins? I guess a pseudo UART device for SSC
isn't enough because bluez needs an alsa device, correct?
For a quick and dirty solution: Is it possible to let bluez reduce the
soundquality so it works over HCI any way?
Are my settings correct or have I missed something?
I know that version 3.36 is a bit old but it works perfectly for me.
Thanks and Best Regards
Sebastian
> test/agent.c seems to be outdated, just as you can see in function
> "static DBusHandlerResult agent_message(...)".
> It is impossible for us to porting python to embedded system, because
> python is too big, and dbus of python needed.
>
> import sys
> import dbus
> import dbus.service
> import dbus.mainloop.glib
Same here
>
> Marcel has said:
> Currently we only have Python scripts test/simple-agent.
> in the future we gonna have a full blown command line client for
BlueZ.
>
> so now bluez 4.28 can run well on embeded system, but cann't run
> pairing.
> let us looking forward to the update of test/agent.c.
Hmm, then i think it would be the best for me to switch back to 3.36 for
the moment until the agent works, or writing my own agent.
Do you think it is possible to write an agent with shell scripts?
That would be an option, but since it is not object oriented it would be
hard I guess.
Regards
Sebastian
> Also when using PCM for SCO then you need to use the D-Bus API for
> headsets and not the ALSA plugin.
>=20
As far as I understand it, D-Bus is for exchanging information between
processes, correct? So I need a device driver and a device where the PCM
stream is sent to. What type of device and driver do I respectively the
D-Bus API need?
Regards
Sebastian
> Python is not really an option for our embedded system.
> I got bluez 4.27 running and all I have tested worked.
> But I have a problem with the pairing:
> Since there is no passkey-agent any longer and I don't want to run
python scripts I tried agent but it starts with an error message:
test/agent.c seems to be outdated, just as you can see in function
"static DBusHandlerResult agent_message(...)".
It is impossible for us to porting python to embedded system, because
python is too big, and dbus of python needed.
import sys
import dbus
import dbus.service
import dbus.mainloop.glib
Marcel has said:
Currently we only have Python scripts test/simple-agent.
in the future we gonna have a full blown command line client for BlueZ.
so now bluez 4.28 can run well on embeded system, but cann't run
pairing.
let us looking forward to the update of test/agent.c.
>=20
> let me repeat this; _NO_ top-posting. I really mean it.
>=20
Sorry, i should have known it.
> > is it possible to use Version 4.27 for embedded systems without any
> > Desktop Environment only with commandline?
> > I will try to switch to 4.27.
>=20
> it is possible. Currently we only have Python scripts that use the
D-Bus
> interfaces, but in the future we gonna have a full blown command line
> client for BlueZ.
Python is not really an option for our embedded system.
I got bluez 4.27 running and all I have tested worked.
But I have a problem with the pairing:
Since there is no passkey-agent any longer and I don't want to run
python scripts I tried agent but it starts with an error message:
=20
/usr/bin # agent 0000
Can't register agent
Method "RegisterAgent" with signature "os" on interface
"org.bluez.Adapter" doesn't exist
It is the new agent:
/usr/bin # agent
Bluetooth agent ver 4.27
Usage:
agent [--device interface] [--path agent-path] <passkey>
var/log/messages says: daemon.err bluetoothd[371]: Failed to access HAL
Maybe this is a problem?
But the device is ok:
~ # hciconfig -a
hci0: Type: UART
BD Address: 00:80:25:07:1F:7E ACL MTU: 310:10 SCO MTU: 64:8
UP RUNNING PSCAN ISCAN AUTH
RX bytes:1161 acl:0 sco:0 events:47 errors:0
TX bytes:502 acl:0 sco:0 commands:47 errors:0
Features: 0xff 0xff 0x8f 0xfe 0x9b 0xf9 0x00 0x80
Packet type: DM1 DM3 DM5 DH1 DH3 DH5 HV1 HV2 HV3
Link policy: RSWITCH HOLD SNIFF PARK
Link mode: SLAVE ACCEPT
Name: 'ravenbt-0'
Class: 0x0a220c
Service Classes: Networking, Capturing
Device Class: Phone, Smart phone
HCI Ver: 2.0 (0x3) HCI Rev: 0x10c6 LMP Ver: 2.0 (0x3) LMP
Subver: 0x10c6
Manufacturer: Cambridge Silicon Radio (10)
Regards
Sebastian
Full var/log/messages:
Dec 31 23:00:08 ravenbt daemon.info bluetoothd[371]: Bluetooth daemon
Dec 31 23:00:08 ravenbt daemon.info bluetoothd[371]: Enabling debug
information
Dec 31 23:00:08 ravenbt daemon.debug bluetoothd[371]: parsing main.conf
Dec 31 23:00:08 ravenbt daemon.debug bluetoothd[371]: discovto=3D0
Dec 31 23:00:08 ravenbt daemon.debug bluetoothd[371]: pairto=3D0
Dec 31 23:00:08 ravenbt daemon.debug bluetoothd[371]: pageto=3D8192
Dec 31 23:00:08 ravenbt daemon.debug bluetoothd[371]: name=3D%h-%d
Dec 31 23:00:08 ravenbt daemon.debug bluetoothd[371]: class=3D0x000100
Dec 31 23:00:08 ravenbt daemon.debug bluetoothd[371]: inqmode=3D0
Dec 31 23:00:08 ravenbt daemon.debug bluetoothd[371]: Key file does not
have key 'DeviceID'
Dec 31 23:00:08 ravenbt daemon.info bluetoothd[371]: Starting SDP server
Dec 31 23:00:08 ravenbt daemon.debug bluetoothd[371]: Loading plugins
/usr/lib/bluetooth/plugins
Dec 31 23:00:08 ravenbt daemon.debug bluetoothd[371]:
register_interface: path /org/bluez/371/any
Dec 31 23:00:08 ravenbt daemon.info bluetoothd[371]: Registered
interface org.bluez.Service on path /org/bluez/371/any
Dec 31 23:00:08 ravenbt daemon.err bluetoothd[371]: Parsing
/etc/bluetooth/network.conf failed: No such file or directory
Dec 31 23:00:08 ravenbt daemon.debug bluetoothd[371]: Config options:
InterfacePrefix=3Dbnep%d, PANU_Script=3D(null), GN_Script=3D(null),
NAP_Script=3D(null), GN_Interface=3Dpan0, NAP_Interface=3Dpan1, =
Security=3Dtrue
Dec 31 23:00:08 ravenbt daemon.err bluetoothd[371]: Can't create GN
bridge
Dec 31 23:00:08 ravenbt daemon.err bluetoothd[371]: Parsing
/etc/bluetooth/input.conf failed: No such file or directory
Dec 31 23:00:08 ravenbt daemon.debug bluetoothd[371]: Unix socket
created: 13
Dec 31 23:00:08 ravenbt daemon.debug bluetoothd[371]: audio.conf: Key
file does not have key 'MaxConnected'
Dec 31 23:00:08 ravenbt daemon.debug bluetoothd[371]: Telephony plugin
initialized
Dec 31 23:00:08 ravenbt daemon.debug bluetoothd[371]: HFP AG features:
"Ability to reject a call" "Enhanced call status" "Extended Error Result
Codes"
Dec 31 23:00:12 ravenbt daemon.info bluetoothd[371]: HCI dev 0
registered
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: child 376 forked
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: child 376 exited
Dec 31 23:00:12 ravenbt daemon.info bluetoothd[371]: HCI dev 0 up
Dec 31 23:00:12 ravenbt daemon.info bluetoothd[371]: Starting security
manager 0
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]:
register_interface: path /org/bluez/371/hci0
Dec 31 23:00:12 ravenbt daemon.info bluetoothd[371]: Registered
interface org.bluez.Service on path /org/bluez/371/hci0
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: proxy_probe: path
/org/bluez/371/hci0
Dec 31 23:00:12 ravenbt daemon.info bluetoothd[371]: Registered
interface org.bluez.SerialProxyManager on path /org/bluez/371/hci0
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]:
network_server_probe: path /org/bluez/371/hci0
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Adding record with
handle 0x10000
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 0000000f-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00000100-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00001002-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00001115-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]:
register_server_record: got record id 0x10000
Dec 31 23:00:12 ravenbt daemon.info bluetoothd[371]: Registered
interface org.bluez.NetworkPeer on path /org/bluez/371/hci0
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]:
network_server_probe: path /org/bluez/371/hci0
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Adding record with
handle 0x10001
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 0000000f-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00000100-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00001002-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00001117-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]:
register_server_record: got record id 0x10001
Dec 31 23:00:12 ravenbt daemon.info bluetoothd[371]: Registered
interface org.bluez.NetworkHub on path /org/bluez/371/hci0
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]:
network_server_probe: path /org/bluez/371/hci0
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Adding record with
handle 0x10002
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 0000000f-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00000100-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00001002-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00001116-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]:
register_server_record: got record id 0x10002
Dec 31 23:00:12 ravenbt daemon.info bluetoothd[371]: Registered
interface org.bluez.NetworkRouter on path /org/bluez/371/hci0
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]:
headset_server_probe: path /org/bluez/371/hci0
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: audio.conf: Key
file does not have key 'Master'
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Adding record with
handle 0x10003
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00000003-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00000100-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00001002-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00001108-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00001112-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00001203-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: a2dp_server_probe:
path /org/bluez/371/hci0
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: audio.conf: Key
file does not have key 'Disable'
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: audio.conf: Key
file does not have group 'A2DP'
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: audio.conf: Key
file does not have group 'A2DP'
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: audio.conf: Key
file does not have group 'A2DP'
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: audio.conf: Key
file does not have group 'A2DP'
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: audio.conf: Key
file does not have key 'Master'
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: SEP 0x375f0
registered: type:0 codec:0 seid:1
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Adding record with
handle 0x10004
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00000019-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00000100-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00001002-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 0000110a-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 0000110d-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]:
avrcp_server_probe: path /org/bluez/371/hci0
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: audio.conf: Key
file does not have key 'Master'
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Adding record with
handle 0x10005
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00000017-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00000100-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00001002-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 0000110c-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 0000110e-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Adding record with
handle 0x10006
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00000017-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00000100-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 00001002-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Record pattern
UUID 0000110e-0000-1000-8000-00805f9
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Creating device
/org/bluez/371/hci0/dev_00_16_38_CA_56_4D
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Creating device
/org/bluez/371/hci0/dev_00_17_E3_94_54_70
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Creating device
/org/bluez/371/hci0/dev_00_1F_01_0D_06_3D
Dec 31 23:00:12 ravenbt daemon.debug bluetoothd[371]: Creating device
/org/bluez/371/hci0/dev_00_13_1E_14_FA_E4
Dec 31 23:00:12 ravenbt daemon.info bluetoothd[371]: Adapter
/org/bluez/371/hci0 has been enabled
Dec 31 23:00:12 ravenbt daemon.err bluetoothd[371]: Failed to access HAL
Hi,
let me repeat this; _NO_ top-posting. I really mean it.
> is it possible to use Version 4.27 for embedded systems without any
> Desktop Environment only with commandline?
> I will try to switch to 4.27.
it is possible. Currently we only have Python scripts that use the D-Bus
interfaces, but in the future we gonna have a full blown command line
client for BlueZ.
Regards
Marcel
Hi Marcel,
is it possible to use Version 4.27 for embedded systems without any
Desktop Environment only with commandline?
I will try to switch to 4.27.
Regards
Sebastian
> -----Original Message-----
> From: Marcel Holtmann [mailto:[email protected]]
> Sent: Sunday, February 01, 2009 5:52 PM
> To: Matuschka, Sebastian
> Cc: [email protected]
> Subject: Re: How can i configure bluez to transmit Voice over PCM
>=20
> Hi Sebastian,
>=20
> > i'm using bluez 3.36 on an embedded system with a CSR Chip
connected.
> > Communication via HCI over UART works well. I need to receive and
send
> > voice from/to a headset and therefore I want to send/get the voice
as
> > PCM to the CSR chip. The PCM pins of the CSR Chip are connected to
the
> > SSC pins of an avr32 microcontroller. The CSR chip is configured to
use
> > PCM as soon as the Host produces Clock, otherwise it tries to
transmit
> > voice per HCI which is too slow (only 115200 baudrate for a 8kHz PCM
> > stream).
> > In etc/bluetooth/audio.conf I have set SCORouting=3DPCM
> > In etc/Bluetooth/asound.conf I have written the following:
> > #####################
> > pcm.headset {
> > type bluetooth
> > }
> > ctl.headset {
> > type bluetooth
> > }
> > #####################
> > But bluez is still sending voice over HCI, which is ok because: How
> > should it know where to transmit the PCM stream. I need a device but
I
> > am not sure what exactly is needed.
> > Do I have to write a dummy soundcard driver which only sends and
> > receives pcm over the ssc pins? I guess a pseudo UART device for SSC
> > isn't enough because bluez needs an alsa device, correct?
> >
> > For a quick and dirty solution: Is it possible to let bluez reduce
the
> > soundquality so it works over HCI any way?
> >
> > Are my settings correct or have I missed something?
>=20
> I would really advise you to start using 4.27 instead of a previous
> major version that is not developed.
>=20
> Also when using PCM for SCO then you need to use the D-Bus API for
> headsets and not the ALSA plugin.
>=20
> Regards
>=20
> Marcel
>=20
Hi Sebastian,
> i'm using bluez 3.36 on an embedded system with a CSR Chip connected.
> Communication via HCI over UART works well. I need to receive and send
> voice from/to a headset and therefore I want to send/get the voice as
> PCM to the CSR chip. The PCM pins of the CSR Chip are connected to the
> SSC pins of an avr32 microcontroller. The CSR chip is configured to use
> PCM as soon as the Host produces Clock, otherwise it tries to transmit
> voice per HCI which is too slow (only 115200 baudrate for a 8kHz PCM
> stream).
> In etc/bluetooth/audio.conf I have set SCORouting=PCM
> In etc/Bluetooth/asound.conf I have written the following:
> #####################
> pcm.headset {
> type bluetooth
> }
> ctl.headset {
> type bluetooth
> }
> #####################
> But bluez is still sending voice over HCI, which is ok because: How
> should it know where to transmit the PCM stream. I need a device but I
> am not sure what exactly is needed.
> Do I have to write a dummy soundcard driver which only sends and
> receives pcm over the ssc pins? I guess a pseudo UART device for SSC
> isn't enough because bluez needs an alsa device, correct?
>
> For a quick and dirty solution: Is it possible to let bluez reduce the
> soundquality so it works over HCI any way?
>
> Are my settings correct or have I missed something?
I would really advise you to start using 4.27 instead of a previous
major version that is not developed.
Also when using PCM for SCO then you need to use the D-Bus API for
headsets and not the ALSA plugin.
Regards
Marcel