[Bug 217931] amd-pstate lacks crucial features: CPU frequency and boost control

[bugzilla-daemon@kernel.org]

https://bugzilla.kernel.org/show_bug.cgi?id=217931

--- Comment #81 from Miroslav Pavleski (miroslav@pavleski.net) ---
> OK great, it looks like the amd-pstate issue is fixed.  I'll include this in
> my next patch series for amd-pstate.  It's good timing because I'm
> overhauling a lot of this code for 6.15.

That is great news! Glad I could be of some use there.


> Unfortunately; your suspend failure is missing the thing I need to tell me
> some important info about the failure mode(the idle mask print).  This is the
> line we should have seen printed.  
>
> https://github.com/torvalds/linux/blob/master/drivers/platform/x86/amd/pmc/pmc.c#L733
> 
> I notice this at the top before your sequence starts though:

>> feb 17 22:06:52 greenstation kernel: NOHZ tick-stop error: local softirq
>> work
>> is pending, handler #08!!!

> And that has me most concerned that the kernel wasn't really ready for
> suspend.  Let's split out your s2idle suspend issue to it's own bug and try
> to figure out the pattern and what debug we'll need.  We can keep this one
> for the amd-pstate issue, and I expect we can land the patch for a later
> 6.14-rc.

Well, I see line 733 calls amd_pmc_idlemask_read which includes the strings
"SMU" and "idlemask" in the output. Unfortunately, idlemask is not found in the
journal.
When grepping for "SMU" I got this:

===
feb 17 22:03:18 greenstation kernel: amdgpu 0000:03:00.0: amdgpu: SMU driver if
version not matched
feb 17 22:03:18 greenstation kernel: amdgpu 0000:03:00.0: amdgpu: SMU is
initialized successfully!
feb 17 22:03:18 greenstation kernel: amdgpu 0000:07:00.0: amdgpu: SMU is
initialized successfully!
feb 17 22:07:09 greenstation kernel: amdgpu 0000:07:00.0: amdgpu: SMU is
resuming...
feb 17 22:07:09 greenstation kernel: amdgpu 0000:07:00.0: amdgpu: SMU is
resumed successfully!
feb 17 22:07:09 greenstation kernel: amdgpu 0000:03:00.0: amdgpu: SMU is
resuming...
feb 17 22:07:09 greenstation kernel: amdgpu 0000:03:00.0: amdgpu: SMU driver if
version not matched
feb 17 22:07:09 greenstation kernel: amdgpu 0000:03:00.0: amdgpu: SMU: I'm not
done with your previous command: SMN_C2PMSG_66:0x00000036
SMN_C2PMSG_82:0x00000000
===

Regarding the NOHZ tick-stop error. I've seen on another thread somewhere that
it is probably related to the external USB-C hub I use. I cannot judge if it is
interfering with the s2idle suspend process.

-- 
You may reply to this email to add a comment.

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