On Sat, 5 Mar 2005, Pavel Machek wrote: > Hi! > > > Now that 2.6.11 is out, we can start to address a number of important > > issues for power management. This overview discusses some of them, > > the things I have run across in my own work. > > Actually pm_message_t conversion is not be complete, yet, and it > probably will not be complete before 2.6.13. Okay, but we can start thinking about the remaining problems and their solutions now. The two most important outstanding issues in the earlier email boil down to this: The need to make sure that suspend/resume are mutually exclusive with probe/release (and possibly other entry points). The USB subsystem has set up its own private locking arrangement to take care of this; perhaps that scheme should be adopted more widely. The need for the sysfs power/state interface to convey more information: driver-specific data, whether or not resume-on- demand should be enabled, ... even something as basic as which the user is trying to do, a suspend or a resume! (If you really want to get fancy, there could be a way for the user to specify that not only the device in question but also all its descendants should be recursively suspended.) Of lesser importance is the question of how to set inactivity timeouts for auto suspend. Once these questions are answered, I think most of the remaining work on suspend/resume can be pushed out to userspace. Alan Stern
