On Sat, 6 Oct 2007, Paul Menage wrote:
> > The getting and putting of the tasks will prevent them from exiting or
> > being deallocated prematurely. But this is also a critical section that
> > will need to be protected by some mutex so it doesn't race with other
> > set_cpus_allowed().
>
> Is that necessary? If some other process calls set_cpus_allowed()
> concurrently with a cpuset cpus update, it's not clear that there's
> any defined serialization semantics that have to be achieved, as long
> as the end result is that the task's cpus_allowed are within the
> cpuset's cpus_allowed.
>
It can race with sched_setaffinity(). It has to give up tasklist_lock as
well to call set_cpus_allowed() and can race
cpus_allowed = cpuset_cpus_allowed(p);
cpus_and(new_mask, new_mask, cpus_allowed);
retval = set_cpus_allowed(p, new_mask);
and allow a task to have a cpu outside of the cpuset's new cpus_allowed if
you've taken it away between cpuset_cpus_allowed() and set_cpus_allowed().
David
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
