Re: Regression: Problem in using USB Bluetooth Dongle (Worked fine in 2.6.24.x)

On Wed, Jun 24, 2009 at 12:14 AM, Marcel Holtmann<marcel@xxxxxxxxxxxx> wrote:
Hi Brian,

At least until kernel version 2.6.24, I was able to use my USB
bluetooth dongle to download contacts from my phone. I made a jump
from 2.6.24 to 2.6.27, so I am not sure exactly which kernel version
starts the problem, but all kernels from the latter (2.6.27) now do
not work. The latest kernel I have tested is 2.6.29.2.

can you test 2.6.30-rc4 or bluetooth-testing.git. I am pretty sure that
whatever broke in 2.6.27 has been fixed by now. Some cheapo USB dongles
behaved wrongly and it was a long time not clear what caused it.

Alas, the regression is still there in 2.6.30-rc5 which I tested over
the weekend.

I also forgot to mention that, even with the kernels with the
regression, device pairing using the dongle still works.

Would a packet capture (usbmon) help? Willing to capture whatever
other debugging data that you think may help

if device pairing works fine, the I guess this is a different problem.
It could be that your USB host controller doesn't like the device. Or
you forgot to load the EHCI driver or something. Otherwise your dongle
should work fine. Since pairing works, what does hciconfig -a says?


EHCI is actually loaded. See my regression report in my original email.

Regarding some conflict with the host controller, well, it did work
fine in 2.6.24.

Is there some other debug output I can collect or some other
information that I can provide to help identify the source of the
regression?


The regression persists even in the latest kernels.

Decided to collect some more debug information on this problem. This
output is from the same system detailed above running on
kernel 2.6.28.10:

the latest kernel is 2.6.30 btw. Can you please test that one.

Will do. Thanks.

Here is the output of hciconfig -a:

hci0: Type: USB
      BD Address: BB:4A:BB:33:40:36 ACL MTU: 339:6 SCO MTU: 180:1
      UP RUNNING PSCAN ISCAN
      RX bytes:14299 acl:0 sco:0 events:386 errors:0
      TX bytes:2221 acl:0 sco:0 commands:231 errors:0
      Features: 0xef 0x3e 0x09 0xf0 0x0b 0x08 0x00 0x80
      Packet type: DM1 DM3 DM5 DH1 DH3 DH5 HV1 HV2 HV3
      Link policy: RSWITCH HOLD SNIFF PARK
      Link mode: SLAVE ACCEPT
      Name: 'oqb-0'
      Class: 0x08010c
      Service Classes: Capturing
      Device Class: Computer, Laptop
      HCI Ver: 1.2 (0x2) HCI Rev: 0x2 LMP Ver: 1.2 (0x2) LMP Subver: 0x2
      Manufacturer: Accel Semiconductor Ltd. (74)
-----------------------------------------------------------------------------

To be honest, I have never seen a chip from this manufacturer. SO I have
no experience with them. Where did you get that dongle from. I might
have to buy one of these.

Here is the output and hcidump of l2ping:

$ sudo l2ping 00:21:AB:9D:0C:C8
Can't connect: Device or resource busy

HCI sniffer - Bluetooth packet analyzer ver 1.40
device: hci0 snap_len: 1028 filter: 0xffffffff
< HCI Command: Create Connection (0x01|0x0005) plen 13
  C8 0C 9D AB 21 00 18 CC 02 00 00 00 01

Please redu them with hcidump -X -V since decoding by hand is way too
painful.

Will do. Thanks.





--
B. Gitonga Marete
Tel: +254-722-151-590
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Relevant Pages