[Public]
Hans,
Do you think you could get one of the folks who reported this regression to
do
a bisect to see which one "fixed" it?
We already know which commit is causing the regression. As Rafael already said the question is why things are not broken in 5.17 and that is not a straight forward bisect. So figuring this out is going to be a lot of work and I'm not sure of that it is worth it. I certainly don't have time to help users with debugging this.
If we get lucky we can come down to some smaller hunks of code that can come back to stable instead of
reverting.
5.17 is almost done and in a couple of weeks Fedora (and Arch and other distros tracking the mainline kernel) will move to 5.17 resolving the wakeup by keyboard issue not working there.
5.16 is not a LTS kernel, so for other distros we would at a minimum figure out what needs to be backported to make things work with 5.15 making the delta / set of possible patches we need even bigger. So as already said IMHO this is not worth it, at least assuming that nothing bad happens when attempting wakeup by keyboard, iow it just does not work and does not put the laptop in some bad state?
Regards,
Hans
You're right, no bad state as a result and probably not worth the tradeoff for a bisect effort. OK, thanks.