> I never tested removing the patch
> (https://github.com/torvalds/linux/commit/4dbd11796f3a8eb95647507befc41995458a4023).
>  Is that something I should do?  I'm happy to do additional testing for a
> permanent solution.

If the quirk patch works for you; I don't see a need to do that.

> So just curious, what are the next steps?  Is this going to be pushed to the
> mainline?

Let me re-open this so we can talk about what to do about it.

The OEM's EC is an opaque black box.
This /seems/ to be a race condition with the Linux kernel and the EC.

* Upstreaming the quirk is one possibility.
This means that every time we see this we add to the quirk list.

* Adding a delay for all hardware is another possibility.  
This seems relatively unfair to other hardware not affected by this issue.

* Look into whether there is anything different in what's sent to the EC
in the delay vs no delay case.  I thought that there shouldn't be
anything different since we're in the s2idle loop, but you can double
check this by:

1) Turning on dynamic debug messages for drivers/acpi/ec.c.
2) Run sequence to reproduce issue on unpatched kernel.
3) Run sequence to reproduce issue on patched kernel.
4) Compare the sequences and see if there are any extra events in the patched 
kernel.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2088733

Title:
  low CPU frequency after wake up AMD Ryzen

To manage notifications about this bug go to:
https://bugs.launchpad.net/linux/+bug/2088733/+subscriptions


-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to