On Fri, 10 Jul 2015, yoma sophian wrote: > hi Alan: > > 2015-05-27 22:40 GMT+08:00 Alan Stern <stern@xxxxxxxxxxxxxxxxxxx>: > > On Wed, 27 May 2015, yoma sophian wrote: > > > >> After reading the kernel power document, freezing-of-tasks.txt , can I > >> get the below conclusion: > >> if I put my thread in freezable, it will get frozen automatically > >> whenever the system goes in to suspend or hibernate. > > > > It's not entirely automatic. Your thread has to call try_to_freeze() > > at appropriate spots. Of course, if your thread is a work queue then > > the workqueue core will do that for you. > after tracing more kernel based your guide, may I get below conclusion: > if I try to create a thread that will do the IO > 1. create the thread, with set_freezable, and call try_to_freeze() at > appropriate spots. > 2. hook my thread on system_freezable_wq > such as calling queue_delayed_work(system_freezable_wq, &mythread, intv) > or mod_delayed_work((system_freezable_wq, &mythread, intv) > > is hub_thread the 1st kind? It depends on which kernel you're looking at. Up until commit 32a6958998c5, which was added in the 3.18 kernel, hub_thread was the 1st kind. In the 3.18 kernel, hub_thread was replaced by a freezable workqueue different from system_freezable_wq. > if Yes, where it try_to_freeze? Before 3.18, try_to_freeze was checked inside wait_event_freezable(), which was called in hub_thread(). > And why not make hub_thread as above #2 kind? We probably could, but it works okay the way it is now. Alan Stern -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
