Re: AMD PMF: CCP PSP fails to reinitialize after hibernation causing TEE errors

[Mario Limonciello <superm1@kernel.org>]

On 10/26/25 6:21 PM, Lars Francke wrote:
> I did try your patch for the TEE enact cmd issue (I have not applied
> the second one yet, I'll try that as well) and I'm sorry to say that
> it didn't help. dmesg looks exactly the same as before as far as I can
> tell.
> 
> I'll triple check if I did everything correctly but I manually
> confirmed that my source tree contains your patch so I'm _fairly_
> certain.
> 
> I did upload the dmesg[1] and more[2] again. Happy to try more patches.
> 
> [1] <https://github.com/lfrancke/amd-pmf-hibernate/blob/main/hibernate-debug-20251027-001557/dmesg-full.log>
> 
> [2] <https://github.com/lfrancke/amd-pmf-hibernate/tree/main/hibernate-debug-20251027-001557>

One way to be sure it's applied is to add a pr_info() message into the 
new function.  Then you'll know exactly when the new function is called 
in the logs.

But assuming that was in place then I feel this will still require some 
other coordination work with the PMF driver then too.
> 
> 
> On Sun, Oct 26, 2025 at 10:16 PM Mario Limonciello <superm1@kernel.org> wrote:
>>
>>
>>
>> On 10/26/25 3:39 PM, Lars Francke wrote:
>>>> Can you try this attached patch which inserts that flow to see if it
>>>> helps?
>>>
>>> I haven't build my own kernel in a good 20 years, I'm on it though.
>>> I've applied it on top of 6.17.5 - is that good or do I need something
>>> newer (6.18 or similar)?
>>>
>>> Thanks!
>>
>> 6.17.5 should be fine.  But I suggest you add both patches.  It would be
>> a nice confirmation if the other patch fixes your power off after
>> hibernate issue.