Hi On Mon, Dec 2, 2013 at 1:24 AM, Anton Vorontsov <anton@xxxxxxxxxx> wrote: > On Mon, Dec 02, 2013 at 01:02:40AM +0100, Michael Trimarchi wrote: >> On Sun, Dec 1, 2013 at 11:37 PM, Anton Vorontsov <anton@xxxxxxxxxx> wrote: >> > On Mon, Nov 25, 2013 at 08:16:34PM +0100, Michael Trimarchi wrote: >> > ... >> >> >> So you can read this value without any type of synchronization >> >> >> with the power_supply_core >> >> >> and sysfs implementation? >> > ... >> >> https://lists.ubuntu.com/archives/kernel-team/2013-January/025206.html >> >> >> >> I found and equivalent scenario that I was trying to said >> > >> > That's a good question, actually. Even though in Pali's case the notifier >> > is atomic (so that we are pretty confident that the object won't be >> > unregistered), there is still a possiblity of a dead lock, for example. So >> >> So if the get_property is a sleeping function it will be a deadlock. Right? > > All kind of bad things might happen, yes. But before that I would expect a > bunch of warnings from might_sleep() stuff. > > I would recommend to test the patches using preempt/smp kernels + various > DEBUG_ kernel options set. > Is it more simple to make it not atomic and use a mutex in get_property? Michael > Thanks, > > Anton -- To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
