FS#68321 - [linux] msr spam in dmesg after upgrading to v5.9.1
Attached to Project:
Arch Linux
Opened by Hubert Jasudowicz (chivay) - Sunday, 18 October 2020, 11:37 GMT
Last edited by Andreas Radke (AndyRTR) - Wednesday, 21 April 2021, 11:17 GMT
Opened by Hubert Jasudowicz (chivay) - Sunday, 18 October 2020, 11:37 GMT
Last edited by Andreas Radke (AndyRTR) - Wednesday, 21 April 2021, 11:17 GMT
|
Details
I've been using
https://github.com/erpalma/throttled
to fix ThinkPad throttling issues.
After upgrading to 5.9.1.arch1-1, I'm being spammed by the msr module every five seconds: [ 1184.272750] msr: Write to unrecognized MSR 0x1a2 by python3 Please report to x86@kernel.org [ 1184.272875] msr: Write to unrecognized MSR 0x1a2 by python3 Please report to x86@kernel.org [ 1184.272971] msr: Write to unrecognized MSR 0x1a2 by python3 Please report to x86@kernel.org [ 1184.273055] msr: Write to unrecognized MSR 0x1a2 by python3 Please report to x86@kernel.org [ 1184.273108] msr: Write to unrecognized MSR 0x1a2 by python3 Please report to x86@kernel.org [ 1184.273145] msr: Write to unrecognized MSR 0x1a2 by python3 Please report to x86@kernel.org [ 1184.273185] msr: Write to unrecognized MSR 0x1a2 by python3 Please report to x86@kernel.org [ 1184.273271] msr: Write to unrecognized MSR 0x1a2 by python3 Please report to x86@kernel.org [ 1184.273353] msr: Write to unrecognized MSR 0x64b by python3 Please report to x86@kernel.org [ 1184.273403] msr: Write to unrecognized MSR 0x64b by python3 Please report to x86@kernel.org [ 1189.278776] filter_write: 22 callbacks suppressed [ 1189.278781] msr: Write to unrecognized MSR 0x1a2 by python3 Please report to x86@kernel.org [ 1189.278874] msr: Write to unrecognized MSR 0x1a2 by python3 Please report to x86@kernel.org [ 1189.278958] msr: Write to unrecognized MSR 0x1a2 by python3 Please report to x86@kernel.org [ 1189.279044] msr: Write to unrecognized MSR 0x1a2 by python3 Please report to x86@kernel.org [ 1189.279100] msr: Write to unrecognized MSR 0x1a2 by python3 Please report to x86@kernel.org [ 1189.279185] msr: Write to unrecognized MSR 0x1a2 by python3 Please report to x86@kernel.org [ 1189.279283] msr: Write to unrecognized MSR 0x1a2 by python3 Please report to x86@kernel.org [ 1189.279366] msr: Write to unrecognized MSR 0x1a2 by python3 Please report to x86@kernel.org [ 1189.279508] msr: Write to unrecognized MSR 0x64b by python3 Please report to x86@kernel.org [ 1189.279559] msr: Write to unrecognized MSR 0x64b by python3 Please report to x86@kernel.org Seems like the fix is already in mainline, however it wasn't pulled into v5.9.1: https://lore.kernel.org/lkml/cover.1598011595.git.chris%40chrisdown.name/ Maybe it's worth applying to Arch kernel. Thanks Hubert Jasudowicz |
This task depends upon
[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=1f35c9c0ce3888405fc813afedaff79de433cf27
[2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=c31feed8461fb8648075ba9b53d9e527d530972f
```
[352325.640829] msr: Write to unrecognized MSR 0x49 by perl
Please report to x86@kernel.org
[352325.644243] msr: Write to unrecognized MSR 0x49 by perl
Please report to x86@kernel.org
[352325.647818] msr: Write to unrecognized MSR 0x49 by perl
Please report to x86@kernel.org
[352325.651859] msr: Write to unrecognized MSR 0x49 by perl
Please report to x86@kernel.org
[352325.655792] msr: Write to unrecognized MSR 0x49 by perl
Please report to x86@kernel.org
[352325.659748] msr: Write to unrecognized MSR 0x49 by perl
Please report to x86@kernel.org
[352325.664735] msr: Write to unrecognized MSR 0x49 by perl
Please report to x86@kernel.org
[352325.669836] msr: Write to unrecognized MSR 0x49 by perl
Please report to x86@kernel.org
[352325.674033] msr: Write to unrecognized MSR 0x49 by perl
Please report to x86@kernel.org
[352325.678639] msr: Write to unrecognized MSR 0x49 by perl
Please report to x86@kernel.org
```
Also reported [2] upstream.
[1] https://github.com/speed47/spectre-meltdown-checker
[2] https://github.com/speed47/spectre-meltdown-checker/issues/385
Not yet; this is the first time I've encountered this-journal dates back to 1 Aug 2020.
I'll give it a shot on my next boot.
The good thing is that right now it didn't affect the operation (yet) of the program.
Kernel parameter `msr.allow_writes=on` (thanks, @jschievink) clears the dmesg entries (for my case).