Messages in this thread |  | | | Date | Tue, 22 Jun 2010 16:34:49 -0400 (EDT) | | From | Alan Stern <> | | Subject | Re: [update] Re: [RFC][PATCH] PM: Avoid losing wakeup events during suspend |
| |
On Tue, 22 Jun 2010, Rafael J. Wysocki wrote:
> Anyway, below's an update that addresses this particular case. > > It adds two more functions, pm_wakeup_begin() and pm_wakeup_end() > that play similar roles to suspend_block() and suspend_unblock(), but they > don't operate on suspend blocker objects. Instead, the first of them increases > a counter of events in progress and the other one decreases this counter. > Together they have the same effect as pm_wakeup_event(), but the counter > of wakeup events in progress they operate on is also checked by > pm_check_wakeup_events(). > > Thus there are two ways kernel subsystems can signal wakeup events. First, > if the event is not explicitly handed over to user space and "instantaneous", > they can simply call pm_wakeup_event() and be done with it. Second, if the > event is going to be delivered to user space, the subsystem that processes > the event can call pm_wakeup_begin() right when the event is detected and > pm_wakeup_end() when it's been handed over to user space.
Or if the event is going to be handled entirely in the kernel but over a prolonged period of time.
> Please tell me what you think.
I like it a lot. It addresses the main weakness in the earlier version. With this, you could essentially duplicate the in-kernel part of the wakelocks/suspend blockers stuff. All except the timed wakelocks -- you might want to consider adding a pm_wakeup_begin_timeout() convenience routine.
Here's another possible enhancement (if you can figure out a way to do it without too much effort): After a suspend begins, keep track of the first wakeup event you get. Then when the suspend is aborted, print a log message saying why and indicating which device was responsible for the wakeup.
One little thing: You have the PCI subsystem call pm_wakeup_event(). If the driver then wants to call pm_wakeup_begin(), the event will get counted twice. I guess this doesn't matter much, but it does seem peculiar.
Alan Stern
|  |