Hello,
As a follow up to the previous Alsa discussion - can anyone tell
me what minimum system required to generate audio out to a headset
is? I have had some success with pulse audio, although not complete
(no audio produced on the headset, although I am able to perform
some system control, such as volume). I was hoping that a program
such as tests/scotest might be sufficient to just transmit raw audio
and have it played on a bluetooth headset.
I've had a look at hcidump output on a working system, and
comparing that to the output of scotest, and the main differences
appear to be some L2CAP/ACL transactions near the beginning:
< ACL data: handle 41 flags 0x02 dlen 18
L2CAP(d): cid 0x0043 len 14 [psm 0]
0000: 1b ef 15 0d 0a 2b 56 47 53 3d 37 0d 0a 8f .....+VGS=7...
< ACL data: handle 41 flags 0x02 dlen 19
L2CAP(d): cid 0x0043 len 15 [psm 0]
0000: 1b ef 17 0d 0a 2b 56 47 4d 3d 31 35 0d 0a 8f .....+VGM=15...
Having a look in audio/headset.c these appear to be setting the gain.
Can anyone tell me how to construct these manually?
I also feel that I'm going about this completely the wrong way -
that there must be an easier way. However given the limited system
that I'm working with bringing in all of DBus + pulse audio just
doesn't seem feasible. Can anyone suggest an alternative method?
Regards,
Andre