On 1 Nov 2018, at 12:00, Omar Sandoval wrote: > On Thu, Nov 01, 2018 at 04:29:48PM +0100, David Sterba wrote: >>> >>> How is that? kthread_stop() frees the task struct, so >>> wake_up_process() >>> would be a use-after-free. >> >> That was an assumption for the demonstration purposes, the wording >> was >> confusing sorry. > > Oh, well in that case, that's exactly what kthread_park() is ;) Stop > the > thread and make wake_up a noop, and then we don't need to add special > cases everywhere else.
This is how Omar won me over to kthread_park(). I think Dave's alternatives can be made correct, but Omar's way solves a whole class of potential problems. The park() makes the thread safely ignore future work, and won't don't have to chase down weird bugs where people are sending future work at the wrong time. -chris
