Messages in this thread | | | Date | Thu, 19 Feb 2015 18:08:24 +0100 (CET) | From | Jiri Kosina <> | Subject | Re: [PATCH 1/3] sched: add sched_task_call() |
| |
On Thu, 19 Feb 2015, Josh Poimboeuf wrote:
> > > > No, these tasks will _never_ make syscalls. So you need to guarantee > > > > they don't accidentally enter the kernel while you flip them. Something > > > > like so should do. > > > > > > > > You set TIF_ENTER_WAIT on them, check they're still in userspace, flip > > > > them then clear TIF_ENTER_WAIT. > > > > > > Ah, that's a good idea. But how do we check if they're in user space? > > > > I don't see the benefit in holding them in a loop - you can just as well > > flip them from the syscall code as kGraft does. > > But we were talking specifically about HPC tasks which never make > syscalls.
Yeah. And getting answer to "is this task_struct currently running in userspace?" is not easy in a non-disruptive way.
Piggy-backing on CONFIG_CONTEXT_TRACKING is one option, but I guess the number of systems that have this option turned on will be rather limited (especially in HPC space).
Sending IPIs around to check whether the task is running in user-space or kernel-space would work, but it's disruptive to the running task, which might be unwanted as well.
-- Jiri Kosina SUSE Labs
| |