You are not logged in.
Downgrading to linux-firmware 20201218.r1803.646f159-1 did not fix the issue for me. Using an AX200 Wifi Card with bluetooth and wireless bluetooth headphones.
Please have a look at my previous post, I'm using the same card so it could possibly solve the issue for you too.
Offline
I have compared all the firmware files that start with "iwl" (even though AX200 ones are only the ones that are prefixed with "iwlwifi-cc-a0") to see if there was any updates between linux-firmware-20201120 and linux-firmware-20210208, and unfortunately there was none.
So I'm not sure the answer here is downgrading the AX200 firmware.
Does anybody know if the bluetooth part maybe uses another firmware file apart from what I compared?
Here is the list if anybody is interested https://docs.google.com/spreadsheets/d/ … TwywDDgBPc
Offline
OK right after I have posted this I have found that the actual bluetooth firmware is "ibt-20" under the "intel" folder. And indeed there is a change on those files.
Offline
Nice job for finding out there was indeed a change, I've been able to find the 2 commits affecting the driver, however, I have no clue where to begin solving/testing/helping to fix the issue.
Kernel commit log
If you take a look at the logs, you can see "Kiran K" committed the changes
If anyone can push me in the direction of fixing this, I'd give it a go...
Offline
You actually pinpointed the correct version. 20201120 is the latest firmware package that AX200 bluetooth works without problems. It's been almost an hour with my test and so far so good.
Offline
Yeah, haven't had a single DC since my post about the version downgrade :knocking-on-wood:
However, it seems like the files that were modified are actually provided by Intel themselves? Can we contact someone to notify them about the issue?
Offline
Indeed the files are provided by Intel and I'm guessing kernel guys just copy it over to their repo. In this case I'm not sure if the bug should be reported to kernel mailing list or to Intel directly.
If you're up to it I would suggest the kernel mail list as I'm pretty sure the developers have a direct line with Intel.
EDIT: The last working version is
FW Build: REL13705
Release Version: 21.120.0.4
Last edited by cb951303 (2021-03-12 20:26:10)
Offline
I sent a mail to the persons involved with the commits (LKML FAQ stated this was the correct approach).
I'll keep you guys posted.
Offline
Unfortunately I had another drop after 2.5 hours of non-stop listening. It's certainly much better compared to before but looks like this firmware also is faulty.
Offline
That would seem very strange. As I've been using my BT headset for a week with 0 issues.
Can you list your Linux related packages to see if there is a mismatch with my versions? (pacman -Qs linux)
Also, have you rebooted your PC after downgrading?
Offline
cars10 wrote:Downgrading to linux-firmware 20201218.r1803.646f159-1 did not fix the issue for me. Using an AX200 Wifi Card with bluetooth and wireless bluetooth headphones.
Please have a look at my previous post, I'm using the same card so it could possibly solve the issue for you too.
Downgrading to linux-firmware-20201120.bc9cd0b-1 didn't work for me. Still having lots of disconnects with Asus Wifi/BT adapter (AX200) and headset LE_WH-1000XM3 (tried on latest kernel versions of linux/linux-lts/linux-tls54).
Offline
I've downgraded my system by following the arch wiki archive guide: https://wiki.archlinux.org/index.php/Ar … cific_date
My mirrorlist file now looks like this (Note: I also had to disable signature checking: https://wiki.archlinux.org/index.php/Pa … _checking):/etc/pacman.d/mirrorlist
Server=https://archive.archlinux.org/repos/2020/12/16/$repo/os/$arch
Note the date I've used is set to "2020/12/16", I previously tried setting it to "2020/12/31" but that didn't solve the issue for me, though it seemed to happen less often... (I suspect that version 20201218... already contained the issue?)!
My linux packages are now the following:local/linux 5.9.14.arch1-1 The Linux kernel and modules local/linux-api-headers 5.8-1 Kernel headers sanitized for use in userspace local/linux-firmware 20201120.bc9cd0b-1 Firmware files for Linux local/linux-headers 5.9.14.arch1-1 Headers and scripts for building modules for the Linux kernel local/linux-lts-headers 5.4.83-1 Headers and scripts for building modules for the LTS Linux kernel
I did this action 3-4 days ago and haven't had a disconnect since.
For what it's worth, I'm using an AX200, and I think 4johnny is correct that the bug goes further back than 20201218. I'm almost a day drop-free after downgrading firmware to 20201120 and LTS kernel.
Offline
Good news, I received feedback already, now it's back to us to help find out more information:
I pushed new firmware for AX200 few days back. Could you please check with that to see if the issue still exists ? For first level of analysis I would need kernel logs (/var/log/syslog). If possible please capture btsnoop logs. You need to start capturing snoop log before the disconnect happens. Please use this command – “sudo btmon -w btsnoop.log”.
I will test this after I finish my working day. If anyone has some spare time, please reply with your logs. The more people that can test it, the faster we may found the solution (or maybe it's already fixed in the latest release!)
Also, if you think the issue seems to be fixed, please let us know.
Thanks in advance!
Offline
I updated to firmware 20210208.b79d239-1 yesterday and haven't dropped since, maybe the issue was resolved by this latest patch?
Offline
The latest patch is added in version 20210315.3568f96-1.
To install this version, you have to manually download the .zst file and install it using
# pacman -U /path/to/package/package_name-version.pkg.tar.zst
Package link: https://archlinux.org/packages/testing/ … -firmware/
ZST file: https://archlinux.org/packages/testing/any/linux-firmware/download
I don't know the specific details why this version is not available through pacman installer...
Previously, I upgraded back to the 20210208 package and it seemed to be working. However, after a day or two, it started happening again...
Offline
The latest patch is added in version 20210315.3568f96-1.
To install this version, you have to manually download the .zst file and install it using
# pacman -U /path/to/package/package_name-version.pkg.tar.zst
Package link: https://archlinux.org/packages/testing/ … -firmware/
ZST file: https://archlinux.org/packages/testing/any/linux-firmware/downloadI don't know the specific details why this version is not available through pacman installer...
Previously, I upgraded back to the 20210208 package and it seemed to be working. However, after a day or two, it started happening again...
Installed 20210315.3568f96-1 an hour ago, rebooted and just got the next random disconnect.
Again with "a2dp-sink 0x55e32a8b5bb8: error 24".
EDIT: Downgrading linux-firmware to 20201120.bc9cd0b-1 works for 2 hours without an issue. Will report on Friday how it went.
EDIT2: Downgrading did not work. Disconnect came after a few minutes. Issue persists.
Last edited by alxndr13 (2021-03-18 08:41:52)
Offline
I was excited to see a new firmware release targeting the AX201 chipset, but after upgrading the firmware to version 20210315.3568f96-1 and rebooting, it only took 13 minutes to disconnect and reproduce the problem.
Second time around, after turning the machine on and off (instead of reboot for good measure), it took 8 minutes to reproduce the issue.
I'm back to using linux-firmware version 20201218.646f159-1, of which I've had zero issues since rolling back the first time.
I'll see about providing some logs via the use of btmon as stated by @4johnny and send them to the author/committer sometime tomorrow morning.
Offline
For everyone's interest, I am also still experiencing the issue on the latest version (20210315.3568f96-1) and send the requested log files to the contact person at Intel.
It seems that both the AX200 and AX201 devices are affected.
As far as I can tell based on your feedback, it looks like the last working versions are:
AX201: 20201218.646f159-1
AX200: 20201120.bc9cd0b-1
Hoping for some good news soon, if you want to share your logs, you can maybe post a pastebin (or similar) link here to your logfile. The log command that's been request is:
sudo btmon -w btsnoop.log
as well as your system log
journalctl -xe > syslog.txt
Last edited by 4johnny (2021-03-18 21:06:25)
Offline
fwiw, there's also a ongoing discussion over there: https://bbs.archlinux.org/viewtopic.php?id=264380
@4Johnny:
I got a recent disconnect including a btsnoop log. https://pastebin.com/BmPXM4vC
journalctl -xe:
Mär 21 00:55:21 bravo pipewire-media-session[1767]: a2dp-sink 0x56269e7f2f98: error 24
Mär 21 00:55:22 bravo bluetoothd[665]: src/profile.c:ext_io_disconnected() Unable to get io data for Hands-Free Voice gateway: getpeername: Transport endpoint is not connected (107)
Mär 21 00:55:22 bravo pipewire-media-session[1767]: Failed to release transport /org/bluez/hci0/dev_60_AB_D2_41_CD_11/sep3/fd0: Method "Release" with signature "" on interface "org.bluez.MediaTransport1" doesn't exist
Mär 21 00:55:22 bravo dbus-daemon[668]: [system] Rejected send message, 0 matched rules; type="method_return", sender=":1.56" (uid=1000 pid=1767 comm="/usr/bin/pipewire-media-session ") interface="(unset)" member="(unset)" error name="(unset)" requested_reply="0" destination=":1.5" (uid=0 pid=665 comm="/usr/lib/bluetooth/bluetoothd ")
Mär 21 00:55:22 bravo bluetoothd[665]: Authentication attempt without agent
Mär 21 00:55:22 bravo bluetoothd[665]: src/profile.c:ext_auth() Hands-Free Voice gateway rejected 60:AB:D2:41:CD:11: org.bluez.Error.Rejected
Last edited by alxndr13 (2021-03-20 23:59:45)
Offline
I'd like to add that I was able to upgrade my firmware to 20210315.3568f96-1 after bluez upgraded to 5.56-2 and I've suffered no problems for 5 hours on PipeWire so far. Got tipped off to that from this thread.
Offline
I'd like to add that I was able to upgrade my firmware to 20210315.3568f96-1 after bluez upgraded to 5.56-2 and I've suffered no problems for 5 hours on PipeWire so far. Got tipped off to that from this thread.
Have the same setup currently, still getting random disconnects. Does not fix it for me. Are you also using the AX200?
Offline
Seems we're having several threads on this issue.
https://bbs.archlinux.org/viewtopic.php?id=264363
"ONLY THOSE WHO ATTEMPT THE IMPOSSIBLE WILL ACHIEVE THE ABSURD"
- Oceania Association of Autonomous Astronauts
Offline
This problem is also occurring on my Dell xps13 (9300) which has the Intel AX1650 chip so I don't think it is limited to AX200 and AX201. Version 20201218.646f159-1 works fine but both 20210208.b79d239-1 and 20210315.3568f96-1 cause drops and automatic reconnects. I have soundcore mini2 speakers connected.
Offline
Damino wrote:I'd like to add that I was able to upgrade my firmware to 20210315.3568f96-1 after bluez upgraded to 5.56-2 and I've suffered no problems for 5 hours on PipeWire so far. Got tipped off to that from this thread.
Have the same setup currently, still getting random disconnects. Does not fix it for me. Are you also using the AX200?
Yes, I'm using the AX200.
Offline
AX200 user here
both linux-firmware 20201120.bc9cd0b-1 and latest one 20210315.r1846.3568f96-1 have disconnects
thats using bluez 5.56-2
Offline