Re: FreeBSD Bluetooth stuff

[Maksim Yevmenkin]

Hello Dustin,

1) compile and install hcidump from the snapshot's ports/ directory
Recompiled and reinstalled.  All went fine.
good

2) make sure you have clean setup, i.e.
  - run # rc.bluetooth stop <device>
  - disconnect the device from the PC
  - reset mouse
  - connect device back to PC
  - run # rc.bluetooth start <device>
  - in separate window run # hcidump -x
  - run # hccontrol -n <device>hci inquiry
  - run # sdptool browse <BD_ADDR_of_your_mouse>
I followed this sequence.  One thing to note is that the mouse tries it's
best to conserve power, so it only responds to any initial queries for a
few seconds after pressing the reset button.  I don't know exactly how
long, but it's pretty short.  Klausler makes a note of this on his site,
so it must have caused him a hiccup or two.
hmmm... if your mouse drains batteries so fast for how you long
can you use it?
Okay, here it is.  Turns out I think my batteries ran out of juice, which
was causing some of the problems.  I wasn't getting any response at all
after all sorts of resets so I tried seeing if I could find the keyboard
and it 'just worked'.  I replaced the batteries and got a response from
hccontrol inquiry.  Just in case replacing the batteries turned out to
reset it more than pressing the button I put the old batteries back in and
still got nothing.  When the batteries 'die' the mouse still lights up
(it's optical), and apparently the radio connection goes first.  Anyway,
without further adue here is the hcidump output:

# hcidump -x
[ inquiry part skipped ]

< HCI Command: Create Connection(0x01|0x0005) plen 13
  A8 D6 7E F2 50 00 18 CC 02 00 C3 4B 01
HCI Event: Command Status(0x0f) plen 4
  00 01 05 04

HCI Event: Connect Complete(0x03) plen 11
  00 29 00 A8 D6 7E F2 50 00 01 00
baseband connection is now open

< HCI Command: Write Link Policy Settings(0x02|0x000d) plen 4
  29 00 0F 00
< ACL data: handle 0x0029 flags 0x02 dlen 12
    L2CAP(s): Connect req: psm 1 scid 0x0040
HCI Event: Number of Completed Packets(0x13) plen 5
  01 29 00 01 00
HCI Event: Command Complete(0x0e) plen 6
  01 0D 08 00 29 00
ACL data: handle 0x0029 flags 0x02 dlen 16
    L2CAP(s): Connect rsp: dcid 0x0040 scid 0x0040 result 1 status 2
this is you PC adjusts link policy settings and request L2CAP
connection on PSM 1 (which is SDP)
ACL data: handle 0x0029 flags 0x02 dlen 16
    L2CAP(s): Connect rsp: dcid 0x0040 scid 0x0040 result 0 status 0
< ACL data: handle 0x0029 flags 0x02 dlen 12
    L2CAP(s): Config req: dcid 0x0040 flags 0x0000 clen 0
HCI Event: Number of Completed Packets(0x13) plen 5
  01 29 00 01 00
ACL data: handle 0x0029 flags 0x02 dlen 14
    L2CAP(s): Config rsp: scid 0x0040 flags 0x0000 result 0 clen 0
ACL data: handle 0x0029 flags 0x02 dlen 16
    L2CAP(s): Config req: dcid 0x0040 flags 0x0000 clen 4
    MTU 48
here your PC tries to configure L2CAP channel (incoming path)
and mouse agrees. after that mouse tries to configure incoming
L2CAP channel and sets incoming MTU to 48 bytes. so far no
problems.
HCI Event: QoS Setup Complete(0x0d) plen 21
  00 29 00 00 01 00 00 00 00 00 00 00 00 F2 2B 00 00 FF FF FF
  FF
oh! i have never seen any device using this :) your mouse
actually wants to set QoS (Quality Of Service). anyway the
status is 00 - success, so no problem here.
< ACL data: handle 0x0029 flags 0x02 dlen 14
    L2CAP(s): Config rsp: scid 0x0040 flags 0x0000 result 0 clen 0
and this is your PC agress to use proposed L2CAP configuration.
So L2CAP connection is open and there was no problem.
< ACL data: handle 0x0029 flags 0x02 dlen 24
    L2CAP(d): cid 0x40 len 20 [psm 1]
        SDP SSA Req: tid 0x0 len 0xf
          pat uuid-16 0x1002 (PubBrwsGrp)
          max 0xffff
          aid(s) 0x0000 - 0xffff
          cont 00
HCI Event: Number of Completed Packets(0x13) plen 5
  01 29 00 01 00
HCI Event: Number of Completed Packets(0x13) plen 5
  01 29 00 01 00
now sdptool sends SDP browse request to your mouse to figure
out which services are available.
ACL data: handle 0x0029 flags 0x02 dlen 14
    L2CAP(d): cid 0x40 len 10 [psm 1]
        SDP SSA Rsp: tid 0x0 len 0x5
          cnt 0x2
 len 0x2 frm->len 0x1 n 0x0
          cont 00
... and the mouse returns nothing back to us :(

< ACL data: handle 0x0029 flags 0x02 dlen 12
    L2CAP(s): Disconn req: dcid 0x0040 scid 0x0040
HCI Event: Number of Completed Packets(0x13) plen 5
  01 29 00 01 00
ACL data: handle 0x0029 flags 0x02 dlen 12
    L2CAP(s): Disconn rsp: dcid 0x0040 scid 0x0040
and at this point sdptool is done and closes the L2CAP
connection.
bottom line: baseband + L2CAP works. you can talk to your
mouse. SDP does not return anything. possible problems:
1) mouse does not fully implement SDP. it is assumed that you have
   to know "magic" PSM for interrupt and control channel on the mouse.
2) SDP interoperability problem

one thing you could try is to ask mouse for the specific service,
rather then for all services. you can do it by
# sdptool search --bdaddr <BD_ADDR> HID

Note: HID is the service you looking for. BD_ADDR is the address
of your mouse.
do you have an open baseband connection? inquiry on the local deivce
will not show remote devices that already connected to the local device.
try to run
# hccontrol -n <device>hci read_connection_list

Yes, this shows I have a connection to my mouse.
that explains why moused does not respond to inquiry anymore.

you can disconnect baseband connection by hand by typing

# hccontol -n <device>hci disconnect <connection_handle>

BTW you can do

# hccontrol help

and

# hccontrol help <command>

for more information

These give a lot more help than the man page.
from the hccontrol(8) man page:

command
             One of the supported commands (see below). Special command
             ``help'' can be used to obtain the list of all supported
             commands. To get more information about specific command
             use ``help command''.
I managed to get 'remote_name_request' to return 'Microsoft Mouse'
(yippee!), but I just arbitrarily chose <ps_rep_mode> and <ps_mode> as 0.
I used 'read_clock_offset' to get the clock offset.  What are these 2
options and how would I properly find them?
well, you have to read Bluetooth spec :) but i can tell you right now
that in this case it is safe to use 0 for all of them :)
Page Scan Mode 0 - is mandatory page scan mode
Clock Offset - it is always safe to set this to 0. If you set it
it *may* speed up page procedure.
perhaps i should change hccontrol(8) so it will use defaults if user
does not provide overrides.
thanks,
max
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message