On Wed, 30 Sep 2015, Rafael J. Wysocki wrote:

From: Rafael J. Wysocki <redacted>

If the platform firmware was involved in the system resume that's
being completed, there is a concern that some devices might have
been reset by it and if those devices had the power.direct_complete
flag set during the preceding suspend transition, they may stay
in a reset-power-on state indefinitely (until they are runtime-resumed
and then suspended again).  That may not be a big deal from the
individual device's perspective, but if the system is an SoC, it may
be prevented from entering deep SoC-wide low-power states on idle
because of that.

To prevent that from happening, force a runtime resume for devices
with power.direct_complete set if the platform firmware was involved
in the resume transition currently in progress.

Something similar was done by the ACPI PM domain, but regardless of
the platform firmware involvement, and the new mechanism should be
sufficient to replace that code, so drop it.

Maybe I'm not reading patch 1/2 correctly, but it looks like an
ordinary ACPI-based desktop PC will always believe the firmware was
involved in an S3 sleep transition.  If that's so then won't this
change defeat all the work being done by people trying to prevent
unneeded runtime resumes during system resume?  direct_complete would 
be useful only on non-ACPI systems.

Alan Stern