On Sat, Feb 27, 2016 at 12:38 AM, Mika Westerberg
[off-list ref] wrote:

On Sat, Feb 27, 2016 at 12:10:03AM -0800, dbasehore . wrote:

quoted

A device is not able to use direct complete if its children do not
also use direct complete. Even though the SCSI layer leaves devices
runtime suspended, the way it does it still prevents its parent from
using direct complete.

Okay.

Do you need to provide ->complete() hook that then resumes the device
when system is resumed (since the device resume hook is never called
when direct_complete is set)? Along the lines of
pm_complete_with_resume_check().

That's an interesting question. Part of direct complete is to leave
the device runtime suspended even after the system resumes if
possible. The comments in pm_complete_with_resume_check indicate that
the firmware may resume a device. I could see this happening with some
kind of SCSI device.

If this is possible, are we able to put the device back into a
consistent state (runtime suspended) by calling suspend for the scsi
device? If not, we might need to use pm_complete_with_resume_check for
the complete callback.