LinuxKPI: pm: add SET_SYSTEM_SLEEP_PM_OPS() and device_can_wakeup()

[bz]

Summary

Add the SET_SYSTEM_SLEEP_PM_OPS() by factoring some other macro code
out in order to set the suspend/resume functions when the struct is
already given. Such is the case in iwlwifi d3.

Also add an initial implementation of device_can_wakeup(). Though
this is likely all we need we have no way of setting the flag for it
yet so leave a pr_debug() and a comment there as well. Until we want
to support WoWLAN thi is likely not needed for wireless.

Both help to make an updated iwlwifi d3 compile.

Sponsored by: The FreeBSD Foundation
MFC after: 3 days

[bz]

sys/compat/linuxkpi/common/include/linux/pm.h

61

This looks odd here but :set list in vi shows tabs.

[bz]

I was about to commit that but then realized I probably cannot? This is needed. I need to bump priority here.
Otherwise I'll change the return into a false, leave the pr_debug() message (adding the review) and remove the struct field for now?

sys/compat/linuxkpi/common/include/linux/pm.h

52

@imp @manu @wulf ; I know Warner has fixed allocations of these changes in the past to get KBI for older versions back.

Here's the problem: struct dev_pm_info is embedded at the end of struct device so this would only "append at the end". Fine.
BUT struct pci_dev has struct device embedded right at the top so all of struct pci_dev is changed.

HOW do we deal with this? How are we ever supposed to change the size of the structs without breaking (external) kmods?