After Linux kernel 6.5 update, AX200 bluetooth error prevents booting

[Macha]

I recently upgraded my 6.5 kernel on KUbuntu 22.04 to kernel 6.5.0-14. On rebooting, I receive the message:

Bluetooth: hci0: Opcode 0xfc1e failed: -16

I know that hci0 is the first (and only) Bluetooth device, which is an Intel AX200 on an MSI MAG x570 Tomahawk Wifi motherboard. I do not understand what the opcode is, or what -16 means here. No amount of effort has gotten my machine to successfully boot, uninterrupted, to a GUI with Bluetooth; I've sort of gotten it to work by disabling it from a command prompt, but this is not a complete solution as it prevents my NVIDIA graphics driver from loading fully (it needs to be uninterrupted from power on, for whatever the reason), limiting me to a very low resolution.

Importantly, when I jump back to kernel 6.2.0+39, everything, including Bluetooth, works perfectly. I've marked this kernel as held, for now, so I don't lose it on an autoremove.

Suspecting that it might just be something odd in Bluetooth firmware memory, I have already powered off my machine, flicked the hard switch off on the PSU, and drained the caps. On booting to 6.5, I now get:

Bluetooth: hci0: Malformed MSFT vendor event: 0x02

However, again, 6.2.0 works perfectly on all accounts, including Bluetooth.

I would very much like to have Bluetooth continue to work on my machine, as while it isn't critical to my multimedia work, I frequently use it to transfer files to wireless devices. Additionally, I find it exceedingly strange that a BT error should prevent booting, but that's Bluetooth for you and I'm sure it has nothing to do with this particular piece of hardware!

What does the above error mean, does anyone understand what's going on here, and is there something I can do to mitigate this situation, aside from waiting for another update to 6.5?

[Macha]

I've encountered at least one friend who has had the same issue with AX200 failures preventing boot, in his case on Ubuntu; and he also uses an Nvidia graphics card. I'm not sure how relevant that is, but the kernel culprit was once again 6.5.0-14; 6.2.0-39 got him to a graphical desktop but, in his case, did not reenable his proprietary graphics driver. I'm not sure how relevant this is, but it doesn't seem to be a just-my-system kind of thing.

If anyone can help us navigate past this issue, or provide any information on the messages above, it would be very much appreciated!

[Alberto_R_Intel]

Macha, Thank you for posting in the Intel® Communities Support.

We will be more than glad to assist you with this matter. 

In order to get the most accurate assistance on this scenario, please follow the instructions in the link below about "Linux* Support for Intel® Wireless Adapters":

https://www.intel.com/content/www/us/en/support/articles/000005511/wireless.html

Any questions, please let me know.

Regards,

Albert R.

Intel Customer Support Technician

[Macha]

Thank you kindly for your prompt response! I do have further questions about these instructions, though.

I see that `iwlwifi-cc-a0-46.ucode` is already installed in `/lib/firmware`. I'm not entirely sure what a `ucode` file is but I'm hesitant to swap it out immediately until I know more; `cmp` detects a difference between the two files on byte 28, but I'm not sure that that's a functional difference.

Reviewing the page, it's suggested to use the firmware git tree to download a version for later kernels. This links to a page of possible ucode downloads, with many of them looking very similar to `iwlwifi-cc-a0-46.ucode` but with a higher number at the end. Are these meant to be more recent versions of the same file?

Lastly, I may note that the provided minimum is listed as working on Kernels 5.1+, which would include 6.5.0; I admit I'm still digging here but that doesn't seem to be true. If this is the same file from default upstream, then it seems to work through 6.2.0-39; 6.5.0-14 seems to break something. (Pending further data of course.)

[Alberto_R_Intel]

Macha, You are very welcome. Thank you very much for your response.

In reference to your inquiries, I will do further research on this subject in order to try to gather the information you are looking for. As soon as I get any updates, I will post all the details on this thread.

Regards,

Albert R.

Intel Customer Support Technician

[Alberto_R_Intel]

Hello Macha, I just received an update on this matter.

While we are still working on this scenario, could you please provide # dmesg > dmesg.log?

Regards,

Albert R.

Intel Customer Support Technician

[Macha]

I'll be happy to. Do you want the current `dmesg`, or the one from booting into recovery mode on 6.5.0-14? I am currently on 6.2.0-39.

[Macha]

I'll send the other one after I've rebooted, started up in advanced mode for 6.5.0-14, and outputted it as a file. Here's 6.2.0-39's dmesg.

[Macha]

You're never going to believe this (I barely do) but after dealing with opcode errors and malformed MSFTs for about a week, suddenly it is successfully...I think... booting to my desktop on 6.5.0-14. Nonetheless, something weird was happening with my Bluetooth and I think this file will still be useful to you. Bluetooth also seems to be functioning suddenly.

(I've encountered other people who have considerable issues with Nvidia hardware as of 6.5.0-14, which is not Intel, but is indicative of something being off here. I have no idea if it's going to sustain this.)

[Macha]

Let it be known that even if we can't resolve what the issue was, I would love it if someone could tell me where I could look up these error codes with my limited knowledge. File transfer over BT is a wonderful thing to have. If I can quickly fix it myself if it happens again in the future, it will be quite nice!

Otherwise, thank you for your attention.

[Alberto_R_Intel]

Macha, You are very welcome. Thank you very much for your reply and for sharing the file.

Perfect, it is great to hear that after dealing with opcode errors and malformed MSFTs the computer is now booting to desktop on 6.5.0-14 with no problems and that also the Bluetooth feature is working correctly as well.

In reference to your question about where to look up the error codes, in that case, we recommend to get in contact directly with your Linux distro for them to provide further assistance and all the details on this subject:

https://www.intel.com/content/www/us/en/support/topics/linux-osv.html

Any other inquiries, do not hesitate to contact us again.

Regards,

Albert R.

Intel Customer Support Technician