Alan Stern <st...@rowland.harvard.edu> writes: [...]
> What we really need to figure out is how to tell the PM core which > devices may safely have their runtime callbacks invoked during system > suspend. For those devices, the core can avoid calling > pm_runtime_disable() during the suspend_late phase. That would address > your requirements, right? Yes, and something I've attempted a few times over the years, most recently during the introduction of the pm_runtime_force* functions[1], which I thought was again an attempt to work around this issue. I don't think Rafael has ever been too thrilled with that idea (including the last time[2]), but I think we're to a point now that we have to manage this somehow. My attempt to let the bus/subsystem/pm_domain set a flag might be too simplistic, but I do agree we do need som way to tell the PM core that runtime PM callbacks are (still) safe. Kevin [1] http://marc.info/?l=linux-pm&m=139343222014989&w=2 [2] http://marc.info/?l=linux-pm&m=139346327619875&w=2 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
