Skip to contentSkip to conversation

New: [REGRESSION] The first attempt to upload firmware for Intel Bluetooth fails

Bug 216936 [email protected] 1 participant 40 messages Jan 15, 2023 – Dec 25, 2025 Bluetooth drivers

lore
Share & tools
lore
#0 lore raw

https://bugzilla.kernel.org/show_bug.cgi?id=216936

            Bug ID: 216936
           Summary: [REGRESSION] The first attempt to upload firmware for
                    Intel Bluetooth fails
           Product: Drivers
           Version: 2.5
    Kernel Version: 6.1.6
          Hardware: AMD
                OS: Linux
              Tree: Mainline
            Status: NEW
          Severity: normal
          Priority: P1
         Component: Bluetooth
          Assignee: [email protected]
          Reporter: [email protected]
        Regression: No

This looks like a new regression for kernel 6.1.6.

Kernel 6.1.4 and previous kernels didn't have the issue.

Hardware: ID 8087:0025 Intel Corp. Wireless-AC 9260 Bluetooth Adapter

From dmesg:

Bluetooth: hci0: FW download error recovery failed (-19)
Bluetooth: hci0: sending frame failed (-19)
Bluetooth: hci0: Failed to read MSFT supported features (-19)

Full log:

Bluetooth: hci0: Found device firmware: intel/ibt-18-16-1.sfi
Bluetooth: hci0: Boot Address: 0x40800
Bluetooth: hci0: Firmware Version: 214-6.22
Bluetooth: hci0: FW download error recovery failed (-19)
Bluetooth: hci0: sending frame failed (-19)
Bluetooth: hci0: Failed to read MSFT supported features (-19)
Bluetooth: hci0: Bootloader revision 0.1 build 42 week 52 2015
Bluetooth: hci0: Device revision is 2
Bluetooth: hci0: Secure boot is enabled
Bluetooth: hci0: OTP lock is enabled
Bluetooth: hci0: API lock is enabled
Bluetooth: hci0: Debug lock is disabled
Bluetooth: hci0: Minimum firmware build 1 week 10 2014
Bluetooth: hci0: Found device firmware: intel/ibt-18-16-1.sfi
Bluetooth: hci0: Boot Address: 0x40800
Bluetooth: hci0: Firmware Version: 214-6.22
Bluetooth: hci0: Waiting for firmware download to complete
Bluetooth: hci0: Firmware loaded in 1111205 usecs
Bluetooth: hci0: Waiting for device to boot
Bluetooth: hci0: Device booted in 14844 usecs
Bluetooth: hci0: Malformed MSFT vendor event: 0x02
Bluetooth: hci0: Found Intel DDC parameters: intel/ibt-18-16-1.ddc
Bluetooth: hci0: Applying Intel DDC parameters completed
Bluetooth: hci0: Firmware revision 0.1 build 214 week 6 2022

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#1 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

https://bugzilla.kernel.org/show_bug.cgi?id=216936

Artem S. Tashkinov ([email protected]) changed:

What |Removed |Added
----------------------------------------------------------------------------

            Summary|[REGRESSION] The first      |First attempt to upload
                   |attempt to upload firmware  |firmware for Intel
                   |for Intel Bluetooth fails   |Bluetooth fails (a timing
                   |                            |issue?)

--- Comment #1 from Artem S. Tashkinov ([email protected]) ---
I see no Bluetooth related changes in 6.1.6 or 6.1.5. This is probably a timing issue but it's still not nice to see.

If there's any additional info I could provide, please let me know.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#2 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

Artem S. Tashkinov ([email protected]) changed:

What |Removed |Added
----------------------------------------------------------------------------
Kernel Version|6.1.6 |6.1.4

--- Comment #2 from Artem S. Tashkinov ([email protected]) --- I had these errors with 6.1.4 as well.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#3 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

The Linux kernel's regression tracker (Thorsten Leemhuis) ([email protected]) changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |[email protected]

--- Comment #3 from The Linux kernel's regression tracker (Thorsten Leemhuis) ([email protected]) --- (In reply to Artem S. Tashkinov from comment #2)

Quoted 1 line

I had these errors with 6.1.4 as well.

But earlier series (e.g. 6.0.y) worked fine?

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#4 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

Mario Limonciello (AMD) ([email protected]) changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |[email protected]

--- Comment #4 from Mario Limonciello (AMD) ([email protected]) ---

Quoted 2 lines

Kernel 6.1.4 and previous kernels didn't have the issue. I had these errors with 6.1.4 as well.

Which one is it?

If you can find a definitive kernel it really was working 100% of the time a bisect would be ideal. But given the ambiguous result you indicated with 6.1.4, I suspect it's not really ever been 100% correct.

Perhaps turning on dynamic debugging for net/bluetooth would be helpful to identify what about the sequence of init causes -ENODEV. Naively it looks to me that the main reason this can get returned is if a command is issued while the device is actively being unregistered.

So perhaps your userspace is racing?

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#5 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #5 from Artem S. Tashkinov ([email protected]) ---
This seemingly happens only after rebooting from Windows 10. It does not happen when cold booting.

I guess if the firmware is already uploaded, then there's an error.

Maybe it happened before, I just didn't reboot after Windows so often.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#6 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

Artem S. Tashkinov ([email protected]) changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |[email protected]

--- Comment #6 from Artem S. Tashkinov ([email protected]) --- Johannes, could you take a look please?

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#7 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

Johannes Berg ([email protected]) changed:

What |Removed |Added
----------------------------------------------------------------------------
CC|[email protected] |

--- Comment #7 from Johannes Berg ([email protected]) --- No, I don't know anything about Bluetooth.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#8 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

James Addison ([email protected]) changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |[email protected]

--- Comment #8 from James Addison ([email protected]) --- I'm seeing a similar problem intermittently, since at least 6.1.0 of the kernel.

Hardware: Intel Corporation Wireless 8260 [8086:24f3]

(yes, it reads like the name a wi-fi card, but it does also contain a bluetooth controller)

Here are the oldest relevant logs that I have available:

Mar 23 21:19:13 kernel: Linux version 6.1.0-6-amd64
([email protected]) (gcc-12 (Debian 12.2.0-14) 12.2.0, GNU ld (GNU
Binutils for Debian) 2.40) #1 SMP PREEMPT_DYNAMIC Debian 6.1.15-1 (2023-03-05)
...
Mar 23 21:19:14 kernel: usb 1-7: USB disconnect, device number 2
Mar 23 21:19:14 kernel: Bluetooth: hci0: Failed to send firmware data (-19)
Mar 23 21:19:14 kernel: Bluetooth: hci0: sending frame failed (-19)
Mar 23 21:19:14 kernel: Bluetooth: hci0: FW download error recovery failed

(-19)

The logs contain 165 occurrences of "hci0: Waiting for firmware download to complete", and 9 occurrences of "hci0: FW download error recovery failed".

I'm not yet sure whether there's a specific series of actions to recreate the failure.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#9 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #9 from James Addison ([email protected]) ---

Quoted 2 lines

I'm not yet sure whether there's a specific series of actions to recreate the failure.

Maybe a clue: of three recent boots with the 8260, two were "cold" (system booted entirely from poweroff state), and one was "warm" (a system reboot). I'm not sure if that terminology is correct, but: the "Failed to send firmware data" error occurred during both of the cold boots, and did not appear during the warm boot.

Does any of that seem relevant in your (9260) case, Artem?

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#10 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #10 from Artem S. Tashkinov ([email protected]) --- I can confirm that firmware gets loaded fine on a cold boot and

I get errors after rebooting from Windows.

This is reproducible in 6.2.10.

Bluetooth: hci0: Found device firmware: intel/ibt-18-16-1.sfi
Bluetooth: hci0: Boot Address: 0x40800
Bluetooth: hci0: Firmware Version: 214-6.22
Bluetooth: hci0: FW download error recovery failed (-19)
Bluetooth: hci0: sending frame failed (-19)
Bluetooth: hci0: Failed to read MSFT supported features (-19)
Bluetooth: hci0: Bootloader revision 0.1 build 42 week 52 2015
Bluetooth: hci0: Device revision is 2
Bluetooth: hci0: Secure boot is enabled
Bluetooth: hci0: OTP lock is enabled
Bluetooth: hci0: API lock is enabled
Bluetooth: hci0: Debug lock is disabled
Bluetooth: hci0: Minimum firmware build 1 week 10 2014
Bluetooth: hci0: Found device firmware: intel/ibt-18-16-1.sfi
Bluetooth: hci0: Boot Address: 0x40800
Bluetooth: hci0: Firmware Version: 214-6.22
Bluetooth: hci0: Waiting for firmware download to complete
Bluetooth: hci0: Firmware loaded in 1104392 usecs
Bluetooth: hci0: Waiting for device to boot
Bluetooth: hci0: Device booted in 14640 usecs
Bluetooth: hci0: Malformed MSFT vendor event: 0x02
Bluetooth: hci0: Found Intel DDC parameters: intel/ibt-18-16-1.ddc
Bluetooth: hci0: Applying Intel DDC parameters completed
Bluetooth: hci0: Firmware revision 0.1 build 214 week 6 2022

BTW "Secure boot is enabled" is wrong. I have it disabled.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#11 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #11 from The Linux kernel's regression tracker (Thorsten Leemhuis) ([email protected]) --- So can anyone reliably reproduce this now? I ask, as I fear this might never
lead to anything unless we find the change where the problem started to happen
using a bisection.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#12 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

espritlibre ([email protected]) changed:

What |Removed |Added
----------------------------------------------------------------------------

                 CC|                            |[email protected]
                   |                            |e

--- Comment #12 from espritlibre ([email protected]) --- i have the same issue

Jul 11 16:06:56 kernel: thinkpad_acpi: rfkill switch tpacpi_bluetooth_sw: radio
is unblocked
Jul 11 16:06:57 kernel: Bluetooth: Core ver 2.22
Jul 11 16:06:57 kernel: NET: Registered PF_BLUETOOTH protocol family
Jul 11 16:06:57 kernel: Bluetooth: HCI device and connection manager
initialized
Jul 11 16:06:57 kernel: Bluetooth: HCI socket layer initialized
Jul 11 16:06:57 kernel: Bluetooth: L2CAP socket layer initialized
Jul 11 16:06:57 kernel: Bluetooth: SCO socket layer initialized
Jul 11 16:06:57 kernel: Bluetooth: hci0: Device revision is 0
Jul 11 16:06:57 kernel: Bluetooth: hci0: Secure boot is enabled
Jul 11 16:06:57 kernel: Bluetooth: hci0: OTP lock is enabled
Jul 11 16:06:57 kernel: Bluetooth: hci0: API lock is enabled
Jul 11 16:06:57 kernel: Bluetooth: hci0: Debug lock is disabled
Jul 11 16:06:57 kernel: Bluetooth: hci0: Minimum firmware build 1 week 10 2014
Jul 11 16:06:57 kernel: Bluetooth: hci0: Bootloader timestamp 2019.40 buildtype
1 build 38
Jul 11 16:06:57 kernel: Bluetooth: hci0: Found device firmware:
intel/ibt-0040-0041.sfi
Jul 11 16:06:57 kernel: Bluetooth: hci0: Boot Address: 0x100800
Jul 11 16:06:57 kernel: Bluetooth: hci0: Firmware Version: 98-13.23
Jul 11 16:06:57 NetworkManager[1454]: <info>  [1689084417.8760] Loaded device
plugin: NMBluezManager
(/usr/lib/NetworkManager/1.42.6-1/libnm-device-plugin-bluetooth.so)
Jul 11 16:06:58 rfkill[1709]: block set for type bluetooth
Jul 11 16:06:58 systemd[1557]: Reached target Bluetooth.
Jul 11 16:06:58 kernel: Bluetooth: hci0: Failed to send firmware data (-19)
Jul 11 16:06:58 kernel: Bluetooth: hci0: sending frame failed (-19)
Jul 11 16:06:58 kernel: Bluetooth: hci0: FW download error recovery failed
(-19)
Jul 11 16:06:58 kernel: Bluetooth: hci0: sending frame failed (-19)
Jul 11 16:06:58 kernel: Bluetooth: hci0: Failed to read MSFT supported features
(-19)
Jul 11 16:06:58 systemd[1]: Starting Bluetooth service...
Jul 11 16:06:58 systemd[1557]: Stopped target Bluetooth.
Jul 11 16:06:58 bluetoothd[1842]: Bluetooth daemon 5.68
Jul 11 16:06:58 systemd[1]: Started Bluetooth service.
Jul 11 16:06:58 systemd[1]: Reached target Bluetooth Support.
Jul 11 16:06:58 systemd[1]: Stopped target Bluetooth Support.
Jul 11 16:06:58 kernel: Bluetooth: BNEP (Ethernet Emulation) ver 1.3
Jul 11 16:06:58 kernel: Bluetooth: BNEP filters: protocol multicast
Jul 11 16:06:58 kernel: Bluetooth: BNEP socket layer initialized
Jul 11 16:06:58 bluetoothd[1842]: Bluetooth management interface 1.22

initialized

i'm not exactly sure when this started happening but i can say it's at least since kernel 6.1.x
i still have these errors on linux-hardened 6.1.38 and linux 6.4.2 and for me it doesn't matter if they appear on a cold or warm boot, for example i can have 3 cold boots in a row without erros and on the fourth cold boot these errors appear.
i can still connect my phone via bluetooth and send files to my phone and back, when i have these errors but whenever i have these errors (most of the time) turning bluetooth on most likely results in a kernel panic and i have to hard power off the laptop which is fscking up my drive (sysrq/reisub ain't working) my wifi card is AX211

i can try to bisect, if someone gives me instructions how to...

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#13 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #13 from Mario Limonciello (AMD) ([email protected]) ---
https://www.kernel.org/doc/html/v6.4/admin-guide/bug-bisect.html

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#14 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #14 from espritlibre ([email protected]) --- i made some progress, i found the kernel version which introduced this bug. good linux-5.19.9.arch1-1
bad linux-6.0.arch1-1

i started the bisect but while building the 2nd kernel i got build errors. i will start from scratch when i find the time again, this already took me a couple hours due to the annoying nature of this bug...

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#15 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #15 from espritlibre ([email protected]) --- sorry! forget what i wrote, 5.19.11 seems to be bad too...

i think this could be a race condition between user space and kernel. because i start my notbeook always in flight mode, in my case it's gnome/wayland. when i turn off flight mode, the bluetooth firmware gets loaded.

if i turn flight mode off and quickly on again, so the firmware download gets interrupted and results in a:
kernel: Bluetooth: hci0: FW download error recovery failed (-19) most of the time

or:
kernel: Bluetooth: hci0: FW download error recovery failed (-108)

i think on boot, the kernel trys to upload the firmware and then the flight mode kicks in, resulting in the error, sometimes. i need to pay more attention and not use flight mode and see if the error appears again.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#16 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #16 from The Linux kernel's regression tracker (Thorsten Leemhuis) ([email protected]) --- (In reply to espritlibre from comment #14)

Quoted 2 lines

i made some progress, i found the kernel version which introduced this bug. good linux-5.19.9.arch1-1

Here you talk about a arch kernel, but...

(In reply to espritlibre from comment #15)

Quoted 1 line

sorry! forget what i wrote, 5.19.11 seems to be bad too...

...here you do not. If that's also an arch kernel and if arch kernels are really close to vanilla, than you want to bisect between the two (unless 5.19.9 turns out to be broken on retesting).

Sight note: it's unlikely that 6.5-rc1 accidentally fixes this, but might be worth a quick test.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#17 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #17 from espritlibre ([email protected]) --- sorry for getting back to this late and the confusion i caused, it's a busy time... the kernels i use are arch kernel, according to the ArchWiki they are pretty much vanilla upstream kernel with some patches applied, but for the current kernel i'm running i see no patches in the PKGBUILD.
i have been observing journal in the and can say that even though i'm booting without being in flightmode but wifi turned on, sometimes

Quoted 8 lines

kernel: Bluetooth: hci0: Failed to send firmware data (-19)
kernel: Bluetooth: hci0: sending frame failed (-19)
kernel: Bluetooth: hci0: FW download error recovery failed (-19)
kernel: Bluetooth: hci0: sending frame failed (-19)
kernel: Bluetooth: hci0: Reading supported features failed (-19)
kernel: Bluetooth: hci0: Error reading debug features
kernel: Bluetooth: hci0: sending frame failed (-19)
kernel: Bluetooth: hci0: Failed to read MSFT supported features (-19)

appears, or:

Quoted 8 lines

kernel: Bluetooth: hci0: Failed to send firmware data (-19)
kernel: Bluetooth: hci0: FW download error recovery failed (-108)
kernel: Bluetooth: hci0: sending frame failed (-19)
kernel: Bluetooth: hci0: sending frame failed (-19)
kernel: Bluetooth: hci0: Reading supported features failed (-19)
kernel: Bluetooth: hci0: Error reading debug features
kernel: Bluetooth: hci0: sending frame failed (-19)
kernel: Bluetooth: hci0: Failed to read MSFT supported features (-19)

i played with bluetooth and the kernel didn't panic.

the last panic i got was 2 days ago and it went like this:
https://0x0.st/H2bz.txt (uploaded for better readability, no javascript needed)

i haven't tried 6.5-rc1, but yesterday i updated to linux-6.4.7.arch1-1. since this bug is so annoying and they happen at random it's hard for me to bisect and i'm not confident in saying kernel X is bad or good. i have set up kdump now, which will automatically dump the crash. i hope this will lead to something useful, i just booted the system 7-8 times but couldn't trigger a panic.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#18 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

Artem S. Tashkinov ([email protected]) changed:

What |Removed |Added
----------------------------------------------------------------------------

             Status|NEW                         |RESOLVED
         Resolution|---                         |CODE_FIX

--- Comment #18 from Artem S. Tashkinov ([email protected]) --- This looks to be fixed in 6.5.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#19 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #19 from James Addison ([email protected]) --- Although I experienced slightly different symptoms -- cold boot firmware downloads failed; warm/reboot firmware downloads succeeded -- when using an 8260 card, I can confirm that the problem appears resolved with a Debian 6.5.0 kernel.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#20 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #20 from Artem S. Tashkinov ([email protected]) --- This is still reproducible in 6.8.5:

dmesg -t | grep -i blue
Bluetooth: Core ver 2.22
NET: Registered PF_BLUETOOTH protocol family
Bluetooth: HCI device and connection manager initialized
Bluetooth: HCI socket layer initialized
Bluetooth: L2CAP socket layer initialized
Bluetooth: SCO socket layer initialized
Bluetooth: hci0: Found device firmware: intel/ibt-18-16-1.sfi
Bluetooth: hci0: Boot Address: 0x40800
Bluetooth: hci0: Firmware Version: 173-5.24
Bluetooth: BNEP (Ethernet Emulation) ver 1.3
Bluetooth: BNEP socket layer initialized
Bluetooth: hci0: FW download error recovery failed (-19)
Bluetooth: hci0: sending frame failed (-19)
Bluetooth: hci0: Reading supported features failed (-19)
Bluetooth: hci0: Error reading debug features
Bluetooth: hci0: HCI LE Coded PHY feature bit is set, but its usage is not supported.
Bluetooth: hci0: sending frame failed (-19)
Bluetooth: hci0: Failed to read MSFT supported features (-19)
Bluetooth: hci0: Bootloader revision 0.1 build 42 week 52 2015
Bluetooth: hci0: Device revision is 2
Bluetooth: hci0: Secure boot is enabled
Bluetooth: hci0: OTP lock is enabled
Bluetooth: hci0: API lock is enabled
Bluetooth: hci0: Debug lock is disabled
Bluetooth: hci0: Minimum firmware build 1 week 10 2014
Bluetooth: hci0: Found device firmware: intel/ibt-18-16-1.sfi
Bluetooth: hci0: Boot Address: 0x40800
Bluetooth: hci0: Firmware Version: 173-5.24
Bluetooth: hci0: Waiting for firmware download to complete
Bluetooth: hci0: Firmware loaded in 1129749 usecs
Bluetooth: hci0: Waiting for device to boot
Bluetooth: hci0: Device booted in 14839 usecs
Bluetooth: hci0: Malformed MSFT vendor event: 0x02
Bluetooth: hci0: Found Intel DDC parameters: intel/ibt-18-16-1.ddc
Bluetooth: hci0: Applying Intel DDC parameters completed
Bluetooth: hci0: Firmware revision 0.1 build 173 week 5 2024
Bluetooth: hci0: HCI LE Coded PHY feature bit is set, but its usage is not supported.
Bluetooth: MGMT ver 1.22
Bluetooth: ISO socket layer initialized
Bluetooth: RFCOMM TTY layer initialized
Bluetooth: RFCOMM socket layer initialized
Bluetooth: RFCOMM ver 1.11

No timings, btintel just tries to load the firmware when it's already there (if you're rebooting from Windows).

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#21 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

Warren Turkal ([email protected]) changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |[email protected]

--- Comment #21 from Warren Turkal ([email protected]) ---
I got a similar issue with btqca. I wonder if this is something bigger than a single module.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#22 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #22 from Artem S. Tashkinov ([email protected]) --- It's not a big issue that's why I kept it close.

I just wonder if it's possible to poll HW and ask it "Hey, do you already have firmware uploaded?" because it surely looks like Linux tries to upload it without much success twice only to fail and realize it's not necessary.

I'm not sure btqca is affected by the same issue and it's not like too many people casually switch between OSes all the time.

At the same time AFAIK ALSA has quite a number of workarounds to deal with the exact same situation: Windows initializes audio HW a certain way and then it starts malfunctioning after rebooting into Linux. It was the case years ago, I've not heard much if any about this recently.

TBO, it more looks like the case of Windows not using cold reboot (and AFAIK by default Linux doesn't do it either) but why would it if e.g. makes booting slower.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#23 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #23 from Warren Turkal ([email protected]) --- AFAIK, no OS uses a cold reboot for the reboot option. Cold reboot implies turning the machine off. Every arch I know of requires a button presses to turn it back on. Very different then just resetting the computer without powering down.

Is there no way to reset the hardware such that it is in the state it is before the power button is pressed?

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#24 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

[email protected] changed:

What |Removed |Added
----------------------------------------------------------------------------

                 CC|                            |cheako+kernel_org@mikemestn
                   |                            |ik.net

--- Comment #24 from [email protected] ---
My testing shows that cold/worm boot is a red herring, the issue is random and if you do it one way enough times you may well find you can get *both* states.

@Mario what were you thinking for DD? For me this happens while going multiuser, so I really would like kernel cmdline or /etc/initramfs setting and not echoing things on a prompt.

I added this to ~/.config/autostart/xmessage.desktop

```desktop
[Desktop Entry]
Type=Application
Exec=sh -c 'sudo dmesg|grep hci0|xmessage -file -'
Hidden=false
X-GNOME-Autostart-enabled=true
Name[en_US]=xmessage
Name=xmessage
Comment[en_US]=xmessage preformance monitoring
Comment=xmessage preformance monitoring
```

I get either no hci0 logs, xmsg is empty, or this:

```dmsg
[    9.536622] Bluetooth: hci0: Bootloader revision 0.3 build 0 week 24 2017
[    9.540760] Bluetooth: hci0: Device revision is 1
[    9.540766] Bluetooth: hci0: Secure boot is enabled
[    9.540769] Bluetooth: hci0: OTP lock is enabled
[    9.540771] Bluetooth: hci0: API lock is enabled
[    9.540773] Bluetooth: hci0: Debug lock is disabled
[    9.540775] Bluetooth: hci0: Minimum firmware build 1 week 10 2014
[    9.544653] Bluetooth: hci0: Found device firmware: intel/ibt-20-1-3.sfi
[    9.544681] Bluetooth: hci0: Boot Address: 0x24800
[    9.544684] Bluetooth: hci0: Firmware Version: 193-33.24
[   10.058240] Bluetooth: hci0: Failed to send firmware data (-19)
[   10.058266] Bluetooth: hci0: sending frame failed (-19)
[   10.058276] Bluetooth: hci0: FW download error recovery failed (-19)
[   10.058313] Bluetooth: hci0: sending frame failed (-19)
[   10.058319] Bluetooth: hci0: Reading supported features failed (-19)
[   10.058322] Bluetooth: hci0: Error reading debug features
[   10.058328] Bluetooth: hci0: HCI LE Coded PHY feature bit is set, but its
usage is not supported.
[   10.058342] Bluetooth: hci0: sending frame failed (-19)
[   10.058349] Bluetooth: hci0: Failed to read MSFT supported features (-19)
```

I had been using BT for a few months, so suddenly loosing it after a kernel upgrade really sucks!

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#25 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #25 from [email protected] --- Created attachment 309051
--> https://bugzilla.kernel.org/attachment.cgi?id=309051&action=edit
Failed to load fw dyndbg.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#26 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #26 from [email protected] --- Created attachment 309052
--> https://bugzilla.kernel.org/attachment.cgi?id=309052&action=edit
Skipped the loading fw attempt

The dyndbg settings are the same, the only change is log ring buffer size... and the contents of that buffer.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#27 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #27 from [email protected] --- Here is my take on setting up dyndbg for this issue.

P.S. Add log_buf_len=4M, to grub cmdline for reboot after the other steps.

Create a file

```
==> /etc/modprobe.d/wifi.conf <==
options iwlwifi dyndbg="+pmfl"
options iwlmvm dyndbg="+pmfl" 
options btusb dyndbg="+pmfl"
options btmtk dyndbg="+pmfl"
options btrtl dyndbg="+pmfl"
options btintel dyndbg="+pmfl" 
options bluetooth dyndbg="+pmfl" 
```

On Debian run update-initramfs -k all -u to include these settings on-boot.

Many of these do not dymdbg on my system.

I'm guessing this line iwlwifi 0000:04:00.0: Detected RF HR B3, rfid=0x10a100 happening early causes the btusb module to be locked out of accessing the device.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#28 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

[email protected] changed:

What |Removed |Added
----------------------------------------------------------------------------

 Attachment #309051|0                           |1

is obsolete| |

--- Comment #28 from [email protected] --- Created attachment 309054
--> https://bugzilla.kernel.org/attachment.cgi?id=309054&action=edit
another attempt, see function trace

Seems like cold boot power-cycles a USB device?

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#31 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

Mark Pearson ([email protected]) changed:

What |Removed |Added
----------------------------------------------------------------------------
CC| |[email protected]

--- Comment #31 from Mark Pearson ([email protected]) --- Bit of a long shot - but try removing the SBDC call in thinkpad_acpi

https://github.com/torvalds/linux/blob/fa084c35afa13ab07a860ef0936cd987f9aa0460/drivers/platform/x86/lenovo/thinkpad_acpi.c#L4181

(afraid you'll need to edit code and build a kernel)

We've seen a few issues on recent platforms with disabling Wifi, related to this call. It's not used on Windows, and it can put the Wifi in a disabled state that it doesn't recover from.

I've been trying to track down what changed and why. We originally thought it was Mediatek modems only, but have seen it on Intel too. On impacted platforms we've been turning that SBDC call into a no-op as a fix in the BIOS (I don't want to break older platforms where this might be needed, so have avoided a kenrnel fix so far).
I'm a bit worried the problem is more widespread than originally expected.

We see the issue when disabling Wifi from the systray on Gnome, but it's fine when done from gnome-control-center. Very weird.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#32 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #32 from [email protected] --- I just added my device to the bt ignore list...
https://github.com/torvalds/linux/blob/master/drivers/platform/x86/lenovo/thinkpad_acpi.c#L4299

from this:

0001C320   86 80 00 00  F3 24 00 00  FF FF FF FF  FF FF FF FF  .....$..........
0001C330   00 00 00 00  00 00 00 00  00 00 00 00  00 00 00 00  ................
0001C340   00 00 00 00  00 00 00 00  86 80 00 00  FD 24 00 00  .............$..
0001C350   FF FF FF FF  FF FF FF FF  00 00 00 00  00 00 00 00  ................
0001C360   00 00 00 00  00 00 00 00  00 00 00 00  00 00 00 00  ................
0001C370   86 80 00 00  26 25 00 00  FF FF FF FF  FF FF FF FF  ....&%..........
0001C380   00 00 00 00  00 00 00 00  00 00 00 00  00 00 00 00  ................
0001C390   00 00 00 00  00 00 00 00  00 00 00 00  00 00 00 00  ................

04:00.0 0280: 8086:2723 (rev 1a)
Subsystem: 8086:0084
Flags: bus master, fast devsel, latency 0, IRQ 69, IOMMU group 12 Memory at c0700000 (64-bit, non-prefetchable) [size=16K]
Capabilities: <access denied>
Kernel driver in use: iwlwifi
Kernel modules: iwlwifi

to this:

0001C320   86 80 00 00  F3 24 00 00  FF FF FF FF  FF FF FF FF  .....$..........
0001C330   00 00 00 00  00 00 00 00  00 00 00 00  00 00 00 00  ................
0001C340   00 00 00 00  00 00 00 00  86 80 00 00  FD 24 00 00  .............$..
0001C350   FF FF FF FF  FF FF FF FF  00 00 00 00  00 00 00 00  ................
0001C360   00 00 00 00  00 00 00 00  00 00 00 00  00 00 00 00  ................
# This entry was only two nibbles off.
0001C370   86 80 00 00  23 27 00 00  FF FF FF FF  FF FF FF FF  ....#'..........
0001C380   00 00 00 00  00 00 00 00  00 00 00 00  00 00 00 00  ................
0001C390   00 00 00 00  00 00 00 00  00 00 00 00  00 00 00 00  ................

and then used modprobe -f to ignore the signature I just broke.

[   88.214055] btusb:btusb_bulk_complete:1520: hci0 urb 000000001b608b6b status
0 count 23
[   88.225307] btusb:btusb_bulk_complete:1520: hci0 urb 0000000022c725e5 status
0 count 23
[  117.676394] stoped logging
[  127.123667] sparse_keymap: bad vermagic: kernel tainted.
[  127.123678] Disabling lock debugging due to kernel taint
[  127.124287] sparse_keymap: module verification failed: signature and/or
required key missing - tainting kernel
[  127.136514] thinkpad_acpi: ThinkPad ACPI Extras v0.26
[  127.136522] thinkpad_acpi: http://ibm-acpi.sf.net/
[  127.136524] thinkpad_acpi: ThinkPad BIOS R0UET78W (1.58 ), EC R0UHT78W
[  127.136528] thinkpad_acpi: Lenovo ThinkPad E585, model 20KV000YUS
[  127.137603] thinkpad_acpi: radio switch found; radios are enabled
[  127.137623] thinkpad_acpi: This ThinkPad has standard ACPI backlight
brightness control, supported by the ACPI video driver
[  127.137628] thinkpad_acpi: Disabling thinkpad-acpi brightness events by
default...
[  127.166948] thinkpad_acpi: Standard ACPI backlight interface available, not
loading native one
[  127.290709] thinkpad_acpi: battery 1 registered (start 75, stop 80,
behaviours: 0xb)
[  127.290735] ACPI: battery: new hook: ThinkPad Battery Extension
[  127.291397] input: ThinkPad Extra Buttons as

/devices/platform/thinkpad_acpi/input/input16

1. Please add a mod option to force this.
2. Add this device.

My dmsgs are still above, but here is more.

Handle 0x000E, DMI type 0, 24 bytes
BIOS Information
Vendor: LENOVO
Version: R0UET78W (1.58 )
Release Date: 11/17/2020
Address: 0xE0000
Runtime Size: 128 kB
ROM Size: 16 MB
Characteristics:
PCI is supported
PNP is supported
BIOS is upgradeable
BIOS shadowing is allowed
Boot from CD is supported
Selectable boot is supported
EDD is supported
3.5"/720 kB floppy services are supported (int 13h) Print screen service is supported (int 5h) 8042 keyboard services are supported (int 9h) Serial services are supported (int 14h) Printer services are supported (int 17h) CGA/mono video services are supported (int 10h) ACPI is supported
USB legacy is supported
BIOS boot specification is supported Targeted content distribution is supported UEFI is supported
BIOS Revision: 1.58
Firmware Revision: 1.58

Handle 0x000F, DMI type 1, 27 bytes
System Information
Manufacturer: LENOVO
Product Name: 20KV000YUS
Version: ThinkPad E585
Wake-up Type: Power Switch
SKU Number: LENOVO_MT_20KV_BU_Think_FM_ThinkPad E585
Family: ThinkPad E585

Handle 0x0010, DMI type 2, 15 bytes
Base Board Information
Manufacturer: LENOVO
Product Name: 20KV000YUS
Version: SDK0J40697 WIN
Asset Tag: Not Available
Features:
Board is a hosting board
Board is replaceable
Location In Chassis: Not Available
Chassis Handle: 0x0000
Type: Motherboard
Contained Object Handles: 0

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#33 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #33 from [email protected] --- Created attachment 309060
--> https://bugzilla.kernel.org/attachment.cgi?id=309060&action=edit
A simple patch, adds my device ID as I suggested.

Sorry for this I know it's etiquette, but I didn't run a linter. So this could be missing a comma or something.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#34 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #34 from [email protected] --- Created attachment 309061
--> https://bugzilla.kernel.org/attachment.cgi?id=309061&action=edit
This patch adds a new parameter, as suggested by myself

Sorry for this I know it's etiquette, but I didn't run a linter. So this could be missing a semicolon or something.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#35 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #35 from Mark Pearson ([email protected]) --- Hi,

Few notes for the patches (though recommend you post these to the platform-driver-x86 mailing list so they can be fully reviewed).

patch1 - put in which platform this is used on in the commit description patch2 - typo in 'btusd'

I want to do some more digging into this problem though.

If this fixes the issue, then you've hit the same issue that is being seen on multiple other newer platforms. The quirk list is OK, but I just worry about side effects without understanding what changed, and why this issue is showing up now - especially as your platform is older.

Mark

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#36 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #36 from [email protected] ---
I'm all for tracking this down... It didn't sit right with me that I can make a call to unplug the bt "dongle", but there isn't an associated call that would plug it back it.

That said, I'm not ready to jump right into building the entire kernel. As indicated a little nocd cracking was all it took to mostly fix this for me.

I'd love a DKMS module that I could use to call these functions from something like sysfs and I'd jump right into documenting what these all do on my platform.

Thanks.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#37 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #37 from James Addison (inactive 20240903) ([email protected]) --- I'm encountering these firmware upload failure messages again with a 6.17.12 kernel (Debian testing/forky 6.17.12+deb14-amd64).

Noting comment 24: this does seem to occur from both cold boot and following a reboot (warm boot).

The hardware I'm using is the same as previously mentioned, and is a wifi+bluetooth controller: Intel Corporation Wireless 8260 [8086:24f3]

Cold boot:

```
[Wed Dec 24 15:30:57 2025] Bluetooth: hci0: Found device firmware:
intel/ibt-11-5.sfi
[Wed Dec 24 15:30:57 2025] Bluetooth: hci0: sending frame failed (-19)
[Wed Dec 24 15:30:57 2025] Bluetooth: hci0: Failed to send firmware data (-19)
[Wed Dec 24 15:31:07 2025] Bluetooth: hci0: FW download error recovery failed
(-110)
[Wed Dec 24 15:31:17 2025] Bluetooth: hci0: Reading supported features failed
(-110)
[Wed Dec 24 15:31:17 2025] Bluetooth: hci0: Error reading debug features
```

Warm boot:

```
[Wed Dec 24 15:32:55 2025] Bluetooth: hci0: Found device firmware:
intel/ibt-11-5.sfi
[Wed Dec 24 15:32:56 2025] Bluetooth: hci0: Failed to send firmware data (-19)
[Wed Dec 24 15:32:56 2025] Bluetooth: hci0: sending frame failed (-19)
[Wed Dec 24 15:32:56 2025] Bluetooth: hci0: FW download error recovery failed
(-19)
[Wed Dec 24 15:32:56 2025] Bluetooth: hci0: sending frame failed (-19)
[Wed Dec 24 15:32:56 2025] Bluetooth: hci0: Reading supported features failed
(-108)
[Wed Dec 24 15:32:56 2025] Bluetooth: hci0: Error reading debug features
```
Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#38 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #38 from [email protected] ---
@James If you didn't see this was an issue with thinkpad_acpi for me, blacklisting fixed it. You could also follow the example in Comment 32 for how to hexedit the module.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.
#39 lore raw

[Bug 216936] First attempt to upload firmware for Intel Bluetooth fails (a timing issue?)

In reply to [email protected]

https://bugzilla.kernel.org/show_bug.cgi?id=216936

--- Comment #39 from James Addison ([email protected]) --- Thanks @cheako; the device-id of my wifi controller is already in the fwbug_cards_ids array, as far as I can tell (0x24F3); I wonder whether the laptop itself (X1 Yoga Gen1) lacks the relevant quirks definition.

Signature · 4 lines
You may reply to this email to add a comment.

You are receiving this mail because:
You are the assignee for the bug.