In D51591#1208761, @bz wrote:

I might go as far as splitting this into (a) the general event handler update and (b) improving linuxkpi?

Superseded by D52861 and D52862

thanks for doing this work!

rebase

More typo fixes and drop __FBSDID tags

Roll typo fixes into this revision

rebase

rebase

thanks for your review!

Remove nhi_compat.h, typo

rebase

Fix S3 suspend.

Rebase on top of D52705.

fixes from review comment in D52598

these changes have been rolled into D52705 because the original commit that had these changes had to be reverted anyways.

In D52598#1201481, @madpilot wrote:

I finally tested this and works fine for me.

I'm sorry I was unable to test it earlier!

In D52598#1201370, @zarychtam_plan-b.pwste.edu.pl wrote:

This patch fixes panic. Please push it when it's ready.

Edit: Unfortunately, suspend is still broken, PC wakes up immediately.

Move struct acpi_wake_prep_context out of acpivar.h

In D52600#1201315, @cy wrote:

This patch fails to address the regression.

Do this correctly.

yeah you're absolutely right, I thought acpi_shutdown's dev was a device but its the actual bus of course so this is easy. Will do that in a minute

Abandoned in favour of the D52036, D52043, D52036 stack.

In D52036#1198932, @markj wrote:

Really just the latter, though "STYPE" is a bit obscure for my taste.

In D52036#1198920, @markj wrote:

I'm not a big fan of the POWER_STYPE_* naming, but hopefully that can be improved and reconciled with the POWER_SLEEP_STATE_* definitions at some point.

forgot to actually do the small fix in question

more small fixes

small fixes

Use own sysctl handler for ACPI stuff so we can still support old S-states here (even though deprecated).

make power_sysctl_stype private

Actually this would need sys/types.h to be included because sysctl.h makes use of u_int and intmax_t. Is it okay to include this?

Accept all USB4 class PCI devices in nhi_pci_probe

Fixes