Bug 194815

You didn't use the FW I attached to https://bugzilla.kernel.org/show_bug.cgi?id=190831 to create this dump and hence the important data is missing.

Also, please mention what Bluetooth device you use.

Created attachment 255125 [details]
New dump and dmesg

(In reply to Emmanuel Grumbach from comment #1)
> You didn't use the FW I attached to
> https://bugzilla.kernel.org/show_bug.cgi?id=190831 to create this dump and
> hence the important data is missing.
> 
> Also, please mention what Bluetooth device you use.

My device: Intel Dual Band Wireless-AC(2*2) 8260, Bluetooth Version 4.1 vPro

I mentioned my device in summary but I don't know where it goes after I post this report.

Sorry for the mistake because I am new to Linux I will try to do it again.

Which means I copy iwlwifi-8000C-27.ucode from bug 190831 to /lib/firmware/iwlwifi-8000c-27.ucode

Before i copy it, my firmware version is: 27.455470.0

after i move your iwlwifi-800c-27 to /lib/firmware
version become to 27.469073.0

(In reply to ahxcker from comment #2)

> My device: Intel Dual Band Wireless-AC(2*2) 8260, Bluetooth Version 4.1 vPro

I think you need to specify an ID of the device. You can use https://linux-hardware.org/index.php?view=howto to upload all info and logs necessary for debugging the problem. And then just attach ID of the probe to this bug instead of attaching all logs separately.

i tr(In reply to Andrey Ponomarenko from comment #3)
> (In reply to ahxcker from comment #2)
> 
> > My device: Intel Dual Band Wireless-AC(2*2) 8260, Bluetooth Version 4.1
> vPro
> 
> I think you need to specify an ID of the device. You can use
> https://linux-hardware.org/index.php?view=howto to upload all info and logs
> necessary for debugging the problem. And then just attach ID of the probe to
> this bug instead of attaching all logs separately.

i tried, it shows me ERROR: can't find 'ifconfig'

i think latest debian use ip tool instant of ifconfig.

What is the Bluetooth device that disconnects? A keyboard? Mouse? What model?

Created attachment 255127 [details]
dmesg for 20170307 12:07

i am using Logitech Mx Master and Bose Soundsport earphone.

Together? Did you try to have only one Bluetooth device at once?

Can you confirm that without wifi it works properly?

(In reply to Emmanuel Grumbach from comment #8)
> Can you confirm that without wifi it works properly?

Yes, after you use command " sudo systemctl stop NetworkManager
mouse working perfectly

Similar issue. 

Intel(R) Dual Band Wireless AC 8265, REV=0x230
Logitech M720 Triathlon

As soon as I start NetworkManager, my mouse is disconnected. Several seconds after I stop NetworkManager and toggle bluetooth on/off, my mouse works again.

Software combination:

Linux 4.10.0
iwlwifi firmware-version: 27.455470.0
Bluetooth: hci0: Firmware revision 0.1 build 82 week 37 2016

Should I create a new bug if it's not the same issue?

No need to create a new bug. 8265 and 8260 are very similar.

Created attachment 255167 [details]
attachment-2064-0.html

There is one thing I want mention, when I use Arch Linux around last
December, my mouse worked fine. Don't know if this is usable.

On Fri, Mar 10, 2017, 12:10 AM <bugzilla-daemon@bugzilla.kernel.org> wrote:

> https://bugzilla.kernel.org/show_bug.cgi?id=194815
>
> --- Comment #11 from Emmanuel Grumbach (emmanuel.grumbach@intel.com) ---
> No need to create a new bug. 8265 and 8260 are very similar.
>
> --
> You are receiving this mail because:
> You are on the CC list for the bug.
> You reported the bug.

(In reply to ahxcker from comment #4)
> i tr(In reply to Andrey Ponomarenko from comment #3)
> > (In reply to ahxcker from comment #2)
> > 
> > > My device: Intel Dual Band Wireless-AC(2*2) 8260, Bluetooth Version 4.1
> > vPro
> > 
> > I think you need to specify an ID of the device. You can use
> > https://linux-hardware.org/index.php?view=howto to upload all info and logs
> > necessary for debugging the problem. And then just attach ID of the probe
> to
> > this bug instead of attaching all logs separately.
> 
> i tried, it shows me ERROR: can't find 'ifconfig'
> 
> i think latest debian use ip tool instant of ifconfig.

Fixed in hw-probe 1.2: https://github.com/linuxhw/hw-probe

Thank you.

(In reply to Emmanuel Grumbach from comment #11)
> No need to create a new bug. 8265 and 8260 are very similar.

Thanks. I can easily reproduce this so I'm happy to help you debug this if you need to test anything.

The firmware team has the data. We'll need to wait for them. Not sure how long it'll take.

Hi,

can you please tell us the exact model of your mouse?
we are going to buy one for testing.

Created attachment 255589 [details]
attachment-7619-0.html

Guys, I don't know what happened. I changed my Debian stretch to arch this
afternoon and it seems there no problem for now.
When I ran arch in last month, I figured out my mouse not work and also my
Bluetooth earphone not work too. Those problems let me changed to Debian
stretch but problem still there.
I don't know what happened today, arch Bluetooth is working. I will try it
tomorrow and give feedback.

On Tue, Mar 28, 2017, 3:25 AM <bugzilla-daemon@bugzilla.kernel.org> wrote:

> https://bugzilla.kernel.org/show_bug.cgi?id=194815
>
> Emmanuel Grumbach (emmanuel.grumbach@intel.com) changed:
>
>            What    |Removed                     |Added
>
> ----------------------------------------------------------------------------
>              Status|ASSIGNED                    |NEEDINFO
>
> --- Comment #16 from Emmanuel Grumbach (emmanuel.grumbach@intel.com) ---
> Hi,
>
> can you please tell us the exact model of your mouse?
> we are going to buy one for testing.
>
> --
> You are receiving this mail because:
> You are on the CC list for the bug.
> You reported the bug.

can you please still tell us the exact model of the mouse?
We want to expand our IOT lab.

thank you.

Created attachment 255595 [details]
Core24 with BT Coex probes enabled

Hello,

the firmware asked me to send you this new firmware for reproduction. It has new debug probes that will help to better understand the issue.

Please send a new dump with this firmware.

Thank you.

Created attachment 255609 [details]
attachment-9750-0.html

Logitech MX.Master
https://www.amazon.com/Logitech-Master-Wireless-Mouse-Computer

On Tue, Mar 28, 2017, 4:10 AM <bugzilla-daemon@bugzilla.kernel.org> wrote:

https://bugzilla.kernel.org/show_bug.cgi?id=194815

--- Comment #18 from Emmanuel Grumbach (emmanuel.grumbach@intel.com) ---
can you please still tell us the exact model of the mouse?
We want to expand our IOT lab.

thank you.

--
You are receiving this mail because:
You are on the CC list for the bug.
You reported the bug.

Created attachment 255611 [details]
attachment-9792-0.html

https://www.amazon.com/Logitech-Master-Wireless-Mouse-Computer/dp/B00TZR3WRM

On Tue, Mar 28, 2017, 10:44 AM <ahxcker@gmail.com> wrote:

> Logitech MX.Master
> https://www.amazon.com/Logitech-Master-Wireless-Mouse-Computer
>
> On Tue, Mar 28, 2017, 4:10 AM <bugzilla-daemon@bugzilla.kernel.org> wrote:
>
> https://bugzilla.kernel.org/show_bug.cgi?id=194815
>
> --- Comment #18 from Emmanuel Grumbach (emmanuel.grumbach@intel.com) ---
> can you please still tell us the exact model of the mouse?
> We want to expand our IOT lab.
>
> thank you.
>
> --
> You are receiving this mail because:
> You are on the CC list for the bug.
> You reported the bug.
>
>

After I change my system to Arch, my Bluetooth become stable.
I tried whole day and only broken connection once.
I still do not have answer for what i did.

You can check the firmware versions of wifi and BT in Debian and arch.

I experience the same problem. My hardware: 
HP Spectre x360 Convertible 13-w0XX
- Intel chipset: 01:00.0 Network controller: Intel Corporation Wireless 8265 / 8275 (rev 78)
- Logitech Mouse MX Master


To temporally solve the connection problem with my mouse I use the work-around described by Ralf: https://www.ralf-oechsner.de/opensource/page/logitech_mx_master

echo "options iwlwifi bt_coex_active=N" | sudo tee -a /etc/modprobe.d/iwlwifi.conf

This works most of the time, but is not stable. When a heavy load occurred on the wifi, the mouse is deactivated. 
Maybe this can brought you further to solve this problem definitive.

Created attachment 255661 [details]
Core24 with BT Coex probes enabled (8265)

What can really help is to provide a firmware dump. For that, I'll need to give you a debug firmware for 8265 which I attached here.

Please follow the instructions here: https://wireless.wiki.kernel.org/en/users/drivers/iwlwifi/debugging#firmware_debugging
You'll need to trigger the dump with:
echo 1 > /sys/kernel/debug/iwlwifi/0000\:0X\:00.0/iwlmvm/fw_dbg_c
(of course, replace X by the appropriate value)

Please take the time to read the privacy notice here:
https://wireless.wiki.kernel.org/en/users/drivers/iwlwifi/debugging#privacy_aspects

Of course, you'll need to remove the iwlwifi setting you added (bt_coex_active=N (and reload iwlwifi or reboot) before collecting the data for us.

Thank you.

The firmware 27 couldn't be leoded by the module. 

At this moment I'm using a Fedora kernel 4.10.5-200.fc25.x86_64, this is for now the latest version. The iwlwifi kernel module is limited to use Firmware iwlwifi-8265-26.ucode

Is there a way to work around by forcing to load the new firmware or can I receive a debug version of the iwlwifi-8265-26.ucode firmware?

(In reply to Kan-Ru Chen from comment #10)
> Similar issue. 
> 
> Intel(R) Dual Band Wireless AC 8265, REV=0x230
> Logitech M720 Triathlon
> 
> As soon as I start NetworkManager, my mouse is disconnected. Several seconds
> after I stop NetworkManager and toggle bluetooth on/off, my mouse works
> again.
> 
> Software combination:
> 
> Linux 4.10.0
> iwlwifi firmware-version: 27.455470.0
> Bluetooth: hci0: Firmware revision 0.1 build 82 week 37 2016

FYI, my current combination works pretty without any coex issue

Linux 4.11.0-rc2
iwlwifi firmware-version: 27.455470.0
Bluetooth: hci0: Firmware revision 0.1 build 103 week 50 2016

(In reply to Bart R. from comment #26)
> The firmware 27 couldn't be leoded by the module. 
> 
> At this moment I'm using a Fedora kernel 4.10.5-200.fc25.x86_64, this is for
> now the latest version. The iwlwifi kernel module is limited to use Firmware
> iwlwifi-8265-26.ucode
> 
> Is there a way to work around by forcing to load the new firmware or can I
> receive a debug version of the iwlwifi-8265-26.ucode firmware?

Sorry, I missed your comment.

You can either install 4.11, or use our backport tree:
http://git.kernel.org/cgit/linux/kernel/git/iwlwifi/backport-iwlwifi.git/

any news here?

Sorry for the delay, I had a busy week. Since the last Fedora update the problems seems to be disappeared. I have also remove the "options iwlwifi bt_coex_active=N" line in /etc/modprobe.d/iwlwifi.conf and the mouse is still working. 

So it looks like that the (Fedora) kernel-core-4.10.9-200.fc25.x86_64 and furthers solves the problem with bluetooth.

what WiFi firmware do you use?

4.10 doesn't use -27.ucode IIRC.

The current firmware that is loaded by the module is: iwlwifi-8265-22.ucode 

apr 19 09:13:51 hp.local kernel: iwlwifi 0000:01:00.0: loaded firmware version 22.391740.0 op_mode iwlmvm
apr 19 09:13:51 hp.local kernel: iwlwifi 0000:01:00.0: Detected Intel(R) Dual Band Wireless AC 8265, REV=0x230

I think that it was not a firmware issue in my case, but more a kernel bug.

(In reply to Bart R. from comment #33)
> I think that it was not a firmware issue in my case, but more a kernel bug.

can't really be.
It can also be a BT firmware change.

I'll close the bug but I'd feel safer if you'd test with -27.ucode