Support for Intel 7260 bluetooth [8087:07dc]

Bug #1209124 reported by Jesse Sung
54
This bug affects 10 people
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
Fix Released
Undecided
Jesse Sung
Nominated for Quantal by Jesse Sung
Nominated for Raring by Jesse Sung
Nominated for Saucy by Jesse Sung

Bug Description

Some 7260 modules are found in the newly released systems. It needs external firmware to work. Support for this module can be found in 3.10 and need to be backported to 3.5 and 3.8.

Jesse Sung (wenchien)
Changed in linux (Ubuntu):
status: New → Fix Released
Revision history for this message
kanor (kanor) wrote :

I have an ux301la laptop with this wireless chip and the bluetooth is not working. Could you help me debugging it? So far I found errors in dmesg:

kanor@kestrel:~$ dmesg | grep -i blue
[ 1.718506] Bluetooth: Core ver 2.16
[ 1.718539] Bluetooth: HCI device and connection manager initialized
[ 1.718546] Bluetooth: HCI socket layer initialized
[ 1.718548] Bluetooth: L2CAP socket layer initialized
[ 1.718556] Bluetooth: SCO socket layer initialized
[ 1.734118] Bluetooth: hci0: read Intel version: 370710018002030d00
[ 1.738944] Bluetooth: hci0: Intel Bluetooth firmware file: intel/ibt-hw-37.7.10-fw-1.80.2.3.d.bseq
[ 3.788882] Bluetooth: BNEP (Ethernet Emulation) ver 1.3
[ 3.788885] Bluetooth: BNEP filters: protocol multicast
[ 3.788892] Bluetooth: BNEP socket layer initialized
[ 3.789654] Bluetooth: RFCOMM TTY layer initialized
[ 3.789664] Bluetooth: RFCOMM socket layer initialized
[ 3.789665] Bluetooth: RFCOMM ver 1.11
[ 3.850669] Bluetooth: hci0 sending Intel patch command (0xfc2f) failed (-19)
[ 5.854659] Bluetooth: hci0 command 0xfc11 tx timeout
[ 13.860301] Bluetooth: hci0 exiting Intel manufacturer mode failed (-110)

kernel is
kanor@kestrel:~$ uname -a
Linux kestrel 3.11.0-14-generic #21-Ubuntu SMP Tue Nov 12 17:04:55 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux

Revision history for this message
Phillip Killewald (paste) wrote :

This bug seems to be affecting me as well. I have the same laptop as kanor mentions, the ASUS UX301LA. When the bluetooth device is blocked via rfkill, the hci0 device shows up properly under hciconfig. However, when the bluetooth device is unblocked, hci0 vanishes.

For example:

$ sudo rfkill block bluetooth
$ sudo rfkill list
0: phy0: Wireless LAN
 Soft blocked: no
 Hard blocked: no
1: asus-wlan: Wireless LAN
 Soft blocked: no
 Hard blocked: no
2: asus-bluetooth: Bluetooth
 Soft blocked: yes
 Hard blocked: no
5: hci0: Bluetooth
 Soft blocked: yes
 Hard blocked: no
$ sudo hciconfig -a
hci0: Type: BR/EDR Bus: USB
 BD Address: 5C:51:4F:0F:42:4E ACL MTU: 1021:5 SCO MTU: 96:5
 DOWN
 RX bytes:1016 acl:0 sco:0 events:104 errors:0
 TX bytes:16335 acl:0 sco:0 commands:103 errors:0
 Features: 0xff 0xfe 0x0f 0xfe 0xdb 0xff 0x7b 0x87
 Packet type: DM1 DM3 DM5 DH1 DH3 DH5 HV1 HV2 HV3
 Link policy: RSWITCH HOLD SNIFF
 Link mode: SLAVE ACCEPT

$ sudo rfkill unblock bluetooth
$ sudo rfkill list
0: phy0: Wireless LAN
 Soft blocked: no
 Hard blocked: no
1: asus-wlan: Wireless LAN
 Soft blocked: no
 Hard blocked: no
2: asus-bluetooth: Bluetooth
 Soft blocked: no
 Hard blocked: no
$ sudo hciconfig -a
[no output]

What's even more interesting is that the device disappears from lsusb output:

$ sudo rfkill block bluetooth
$ sudo lsusb -d 8087:07dc
Bus 002 Device 007: ID 8087:07dc Intel Corp.
$ sudo rfkill unblock bluetooth
$ sudo lsusb -d 8087:07dc
[no output]

Relevant information:

$ dmesg | grep -i bluetooth
[ 9.128331] Bluetooth: Core ver 2.16
[ 9.128723] Bluetooth: HCI device and connection manager initialized
[ 9.128731] Bluetooth: HCI socket layer initialized
[ 9.128734] Bluetooth: L2CAP socket layer initialized
[ 9.128741] Bluetooth: SCO socket layer initialized
[ 9.135096] Bluetooth: BNEP (Ethernet Emulation) ver 1.3
[ 9.135100] Bluetooth: BNEP filters: protocol multicast
[ 9.135107] Bluetooth: BNEP socket layer initialized
[ 9.143738] Bluetooth: RFCOMM TTY layer initialized
[ 9.143751] Bluetooth: RFCOMM socket layer initialized
[ 9.143753] Bluetooth: RFCOMM ver 1.11
[ 10.067400] Bluetooth: hci0: read Intel version: 370710018002030d2e
[ 10.067404] Bluetooth: hci0: Intel device is already patched. patch num: 2e
[ 1112.489206] Bluetooth: hci0: read Intel version: 370710018002030d00
[ 1112.493990] Bluetooth: hci0: Intel Bluetooth firmware file: intel/ibt-hw-37.7.10-fw-1.80.2.3.d.bseq
[ 1112.645191] Bluetooth: hci0: Intel Bluetooth firmware patch completed and activated

$ uname -a
Linux zadkiel 3.11.0-15-generic #23-Ubuntu SMP Mon Dec 9 18:17:04 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux

Revision history for this message
Anton (o-anton-g) wrote :

Hi, I see a fixed has been released for this issue. Where can I find this fix?

I have an Asus UX301LAA, where I can not get the Bluetooth to work.

$uname -a
Linux anton-UX301LAA 3.13.0-36-generic #63-Ubuntu SMP Wed Sep 3 21:30:07 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux

Revision history for this message
John Wang (johnwang) wrote :

@Anton:

The fix that was released is the addition of Intel 7260 Bluetooth firmware for device ID 8087:07dc, which has been shipping in Ubuntu since at least Saucy.

It doesn't mean that Bluetooth is fixed. Bluez (and/or the kernel's Bluetooth stack) still doesn't handle the setup properly because, thanks to bad design by ASUS, the Bluetooth hardware is poorly integrated on the UX301 models. See:

http://askubuntu.com/questions/396021/what-is-causing-my-intel-7260-bluetooth-device-to-disconnect-when-i-unblock-it-w

The workaround given in the first post of the above link enables Bluetooth, but at the cost of disabling USB 3.0. Another workaround, which was a pain to set up and only gave me occasional success, is to uninstall the Bluez stack and replace it with the Blueman stack, but I really don't recommend that (I've switched back to Bluez).

Revision history for this message
Chronos (chronos00) wrote :

Is there any work being done from the Linux community to get the bluetooth module for the UX301LA working with xHCI?

I guess we could also contact ASUS and demand a BIOS update from them. If so, what exactly should we be telling them?

Thank you.

Revision history for this message
Chronos (chronos00) wrote :

By the way, I just confirmed that disabling "asus-nb-wmi" allows me to bring hci0 up and use the bluetooth module.

The command sequence is the following:
`sudo rmmod asus-nb-wmi`
`sudo rfkill unblock bluetooth` (this would usually completely disconnect the bluetooth module)
`sudo hciconfig hci0 up`

At this time, you can use Bluez normally. Of course, the side effect of removing "asus-nb-wmi" is that I lose all the "function-keys" (Brightness control, volume control, etc). As soon as I bring this module back up with modprobe, the bluetooth module is disconnected again.

I got this information from this comment at ask-Ubuntu:
http://askubuntu.com/a/541546/25561

Does this mean that we should file a bug report for asus-nb-wmi module?

Revision history for this message
John Wang (johnwang) wrote :

@Chronos:

Our issue is distinct from this bug. I intend to file a bug upstream (kernel.org) once I finish working out how the ASUS platform driver handles Bluetooth and I have a reasonable idea to whom the bug report should be directed. I'll post a link back here when that's done.

In the meantime, I've discovered the least intrusive (but still dirty) workaround yet:

1. In Unity's Bluetooth indicator, flip the "Bluetooth" toggle switch to "ON". Because of our bug, this causes the "hci0" Bluetooth device to be unregistered from the kernel's rfkill subsystem (see "rfkill list all").

2. In a terminal do "sudo rfkill block N", where N is the ID number of the remaining "asus-bluetooth" WMI device as given by "rfkill list all". (In my case it's "sudo rfkill block 2".) This causes the "hci0" device to be registered with rfkill again, and it will be soft-unblocked. In the Bluetooth indicator menu you should now see both "Bluetooth" and "Visible" toggled to "ON".

3. Connect Bluetooth devices normally, but be careful; *don't* manually flip the "Bluetooth" or "Visible" switches, whether in the Bluetooth indicator or in "System Settings -> Bluetooth", because that causes the "hci0" adapter to disappear again.

Revision history for this message
Chronos (chronos00) wrote :

@John Wang

Thank you for your workaround. I can confirm this worked for me as well (comment #7).

The fact that I can't find a way to make this hack automatic or survive reboots forces me to sacrifice an USB port for a small external bluetooth dongle. A pity for such an expensive laptop...

Lets hope we can solve the bug quickly. I offer myself to run any tests that are needed.

Do you have a link to the upstream bug report?

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.