Public bug reported:
**Title:** Bluetooth HFP microphone extremely quiet on Intel Wireless
8260; kernel reports corrupted SCO packets and unknown SCO connection
handle 257
### System
- OS: Ubuntu 26.04.1
- Kernel tested: `7.0.0-34-generic` and `7.0.0-38-generic`
- Bluetooth controller: Intel Wireless 8260, PCI ID `8086:24f3`, revision `3a`
- Bluetooth driver modules: `btusb`, `btintel`, `bluetooth`
- BlueZ: `5.85-4ubuntu0.4`
- PipeWire: `1.6.2`
- WirePlumber: `0.5.13`
- Firmware package: `linux-firmware-intel-wireless`
- Controller firmware reports: `Firmware revision 0.0 build 14 week 44 2021`
- Headset: Bowers & Wilkins Px8
### Problem
The Px8 microphone works normally when paired with a smartphone, but its
microphone input is extremely quiet on Ubuntu. Bluetooth playback works
normally.
The issue is reproducible with direct PipeWire recording and is not
limited to a particular speech-recognition application.
### Measurements
Using `ffmpeg`'s `volumedetect` filter on recorded WAV files:
- Px8 HFP/mSBC recording: mean volume between approximately `-50.4 dB` and
`-55.0 dB`; latest peak `-23.5 dB`.
- Internal laptop microphone: mean volume approximately `-4.2 dB` to `-9.2 dB`.
Switching between mSBC and CVSD did not materially improve the Px8
microphone level. Disconnecting and reconnecting the headset also did
not solve the issue.
The Bluetooth input source was not muted and its configured volume was
approximately 96%.
### Kernel messages during microphone recording
The following messages appeared while recording from the Px8 microphone:
Bluetooth: hci0: corrupted SCO packet
Bluetooth: hci0: SCO packet for unknown connection handle 257
The `corrupted SCO packet` message appeared repeatedly. The unknown-
connection-handle message appeared twice in the captured excerpt.
### Troubleshooting already performed
- Confirmed that the headset microphone works normally with a smartphone.
- Tested both `headset-head-unit` (mSBC) and `headset-head-unit-cvsd` (CVSD).
- Measured recordings directly with PipeWire rather than relying only on the
GNOME input meter.
- Reconnected the headset and repeated the recording test.
- Confirmed that PipeWire and WirePlumber were running.
- Tested kernels `7.0.0-34-generic` and `7.0.0-38-generic`.
- Confirmed that the `btusb` and `btintel` kernel modules are loaded.
- Checked that `/usr/lib/firmware/intel/ibt-11-5.ddc.zst` and
`/usr/lib/firmware/intel/ibt-11-5.sfi.zst` exist and belong to
`linux-firmware-intel-wireless`.
- No firmware files have been manually modified.
### Related reports
Two upstream reports describe similar SCO transport symptoms:
1. An Intel AX201/AX211 report describes `SCO packet for unknown connection
handle 257` after HFP profile changes, in a particular Linux/WirePlumber
version combination.
2. A WirePlumber report describes corrupted SCO packets and an unknown
connection handle with Bluetooth HFP audio failures.
The hardware and software versions differ from this report, so the
underlying cause may not be the same.
### Questions
1. Do the SCO errors indicate a Bluetooth controller/firmware problem, a Linux
Bluetooth SCO connection-state issue, or an interaction with
PipeWire/WirePlumber?
2. Is there a known fix applicable to Intel Wireless 8260 controllers?
3. Would an HCI snoop trace or additional kernel diagnostics help distinguish a
firmware problem from a SCO connection-state problem?
4. Is there a recommended regression test for the Intel 8260 with Ubuntu 26.04
and kernel 7.0?
## 1. System configuration
- Operating system: Ubuntu 26.04.1
- Laptop: ThinkPad T460
- Kernel versions tested: `7.0.0-34-generic` and `7.0.0-38-generic`
- Bluetooth/WLAN adapter: Intel Wireless 8260
- PCI ID: `8086:24f3`, revision `3a`
- Bluetooth kernel modules loaded: `btusb`, `btintel`, `bluetooth`
- BlueZ: `5.85-4ubuntu0.4`
- PipeWire: `1.6.2`
- WirePlumber: `0.5.13`
- Headset: Bowers & Wilkins Px8
- Intel Bluetooth firmware reported by the kernel: `Firmware revision 0.0 build
14 week 44 2021`
- Firmware package: `linux-firmware-intel-wireless`
- Relevant firmware files present:
- `/usr/lib/firmware/intel/ibt-11-5.ddc.zst`
- `/usr/lib/firmware/intel/ibt-11-5.sfi.zst`
## 2. Problem description
The Bowers & Wilkins Px8 microphone works normally when connected to a
smartphone. On Ubuntu, Bluetooth playback works, but the headset
microphone input is extremely quiet.
The problem is reproducible in direct PipeWire recordings. The internal
laptop microphone records at a much higher level.
The issue persists after reconnecting the headset and after switching
between the mSBC and CVSD headset profiles.
## 3. Recording measurements
Measurements were obtained from WAV recordings using FFmpeg's
`volumedetect` filter.
| Recording source/profile | Mean volume | Peak volume |
|---|---:|---:|
| Px8, mSBC | -50.4 dB | -27.6 dB |
| Px8, mSBC | -52.6 dB | -27.1 dB |
| Px8, mSBC, after reconnecting | -55.0 dB | -23.5 dB |
| Px8, CVSD | -49.2 dB | -31.0 dB |
| Internal microphone, recording 1 | -9.2 dB | 0.0 dB |
| Internal microphone, recording 2 | -4.2 dB | 0.0 dB |
The headset input is therefore approximately 40–50 dB lower in mean
volume than the internal microphone in the recorded samples. The
measurements depend on the speech and recording conditions, but the
difference is substantial.
The Px8 PipeWire input source was not muted. Its configured source
volume was approximately 96% (`-1.19 dB`).
## 4. Kernel messages observed during microphone recording
The following messages appeared during an active recording test:
Bluetooth: hci0: corrupted SCO packet
The message appeared repeatedly.
The following message also appeared:
Bluetooth: hci0: SCO packet for unknown connection handle 257
This appeared twice in the captured excerpt.
These messages occurred while testing the Px8 microphone. Their causal
relationship to the low recording level has not yet been established.
## 5. Troubleshooting already performed
1. Confirmed that the Px8 microphone works normally with a smartphone.
2. Confirmed that Bluetooth audio playback works on Ubuntu.
3. Compared the headset microphone with the internal laptop microphone.
4. Recorded microphone samples directly through PipeWire and measured their
levels using FFmpeg.
5. Tested the `headset-head-unit` profile using mSBC at 16 kHz.
6. Tested the `headset-head-unit-cvsd` profile using CVSD at 8 kHz.
7. Neither profile produced a material improvement in microphone level.
8. Disconnected and reconnected the headset, then repeated the recording test.
9. Confirmed that PipeWire and WirePlumber were running.
10. Confirmed that the Bluetooth input source was not muted and was set to
approximately 96% volume.
11. Tested kernels `7.0.0-34-generic` and `7.0.0-38-generic`. The microphone
issue was still observed with both, although the tests were not fully
controlled under identical conditions.
12. Confirmed that the `btusb`, `btintel` and `bluetooth` kernel modules were
loaded.
13. Checked the USB runtime power-management setting for the Bluetooth device
and temporarily set it to `on`. This did not resolve the microphone issue.
14. Confirmed that the firmware files `ibt-11-5.ddc.zst` and `ibt-11-5.sfi.zst`
are installed and belong to the Ubuntu package `linux-firmware-intel-wireless`.
15. Checked the kernel log for firmware-related messages. The available excerpt
reported firmware build 14, week 44 of 2021; no confirmed DDC loading error was
identified in that excerpt.
16. Checked the installed firmware file's SHA-256 hash. No authoritative
comparison has yet established whether the file is defective.
## 6. Questions for Ubuntu kernel/Bluetooth maintainers
1. Can the repeated `corrupted SCO packet` and `SCO packet for unknown
connection handle 257` messages indicate a known Intel Bluetooth controller,
firmware, or SCO connection-state problem?
2. Is there a known issue affecting Intel Wireless 8260 controllers on Ubuntu
26.04 or Linux kernel 7.0?
3. Could this issue cause an extremely low microphone recording level without
completely disabling HFP audio?
4. Would an HCI snoop trace or additional kernel diagnostics help determine
whether the SCO transport is corrupted or the connection state becomes
inconsistent?
5. Is there a recommended regression test or a known fix for this
hardware/software combination?
ProblemType: Bug
DistroRelease: Ubuntu 26.04
Package: linux-image-7.0.0-34-generic 7.0.0-34.34
ProcVersionSignature: Ubuntu 7.0.0-34.34-generic 7.0.14
Uname: Linux 7.0.0-34-generic x86_64
ApportVersion: 2.34.1-0ubuntu0.1
Architecture: amd64
AudioDevicesInUse:
USER PID ACCESS COMMAND
/dev/snd/controlC0: ulrich 3400 F.... wireplumber
/dev/snd/seq: ulrich 3381 F.... pipewire
CasperMD5CheckResult: unknown
CurrentDesktop: ubuntu:GNOME
Date: Thu Oct 22 11:56:01 2026
HibernationDevice: RESUME=UUID=5a99a614-ef51-4b2f-9b3c-f73d23cfd8d6
InstallationDate: Installed on 2016-07-22 (3744 days ago)
InstallationMedia: Ubuntu 16.04.1 LTS "Xenial Xerus" - Release amd64 (20160719)
Lsusb:
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 001 Device 002: ID 8087:0a2b Intel Corp. Bluetooth wireless interface
Bus 001 Device 003: ID 138a:0017 Validity Sensors, Inc. VFS 5011 fingerprint
sensor
Bus 001 Device 004: ID 04ca:7058 Lite-On Technology Corp. Integrated Camera
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
MachineType: LENOVO 20FMS03600
ProcFB: 0 i915drmfb
ProcKernelCmdLine: BOOT_IMAGE=/vmlinuz-7.0.0-34-generic
root=/dev/mapper/ubuntu--vg-root ro quiet splash
crashkernel=2G-4G:320M,4G-32G:512M,32G-64G:1024M,64G-128G:2048M,128G-:4096M
PulseList: Error: command ['pacmd', 'list'] failed with exit code 1: No
PulseAudio daemon running, or not running as session daemon.
SourcePackage: linux
UpgradeStatus: Upgraded to resolute on 2026-09-29 (23 days ago)
dmi.bios.date: 03/23/2016
dmi.bios.release: 1.10
dmi.bios.vendor: LENOVO
dmi.bios.version: R06ET36W (1.10 )
dmi.board.asset.tag: Not Available
dmi.board.name: 20FMS03600
dmi.board.vendor: LENOVO
dmi.board.version: Not Defined
dmi.chassis.asset.tag: No Asset Information
dmi.chassis.type: 10
dmi.chassis.vendor: LENOVO
dmi.chassis.version: None
dmi.ec.firmware.release: 1.8
dmi.modalias:
dmi:bvnLENOVO:bvrR06ET36W(1.10):bd03/23/2016:br1.10:efr1.8:svnLENOVO:pn20FMS03600:pvrThinkPadT460:rvnLENOVO:rn20FMS03600:rvrNotDefined:cvnLENOVO:ct10:cvrNone:skuLENOVO_MT_20FM_BU_Think_FM_ThinkPadT460:pfaThinkPadT460:
dmi.product.family: ThinkPad T460
dmi.product.name: 20FMS03600
dmi.product.sku: LENOVO_MT_20FM_BU_Think_FM_ThinkPad T460
dmi.product.version: ThinkPad T460
dmi.sys.vendor: LENOVO
** Affects: linux (Ubuntu)
Importance: Undecided
Status: New
** Tags: amd64 apport-bug resolute wayland-session
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2171921
Title:
Bluetooth HFP microphone extremely quiet on Intel Wireless 8260;
kernel reports corrupted SCO packets and unknown SCO connection handle
257
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2171921/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
