Re: [rfc] "fair" rw spinlocks

From: Peter Zijlstra
Date: Mon Dec 07 2009 - 15:39:11 EST


On Mon, 2009-12-07 at 19:32 +0100, Oleg Nesterov wrote:
> On 12/05, Eric W. Biederman wrote:
> >
> > Atomically sending signal to every member of a process group, is the
> > big fly in the ointment I am aware of. Last time I looked I could
> > not see how to convert it rcu.
>
> I am not sure, but iirc we can do this lockless (under rcu_lock).
> We need to modify pid_link to use list_entry and attach_pid() should
> add the new task to the end. Of course we need more changes, but
> (again iirc) this is not too hard.
>
> > This is a pain because we occasionally signal a process group from
> > interrupt context.
>
> Only send_sigio/etc does so, right?
>
>
> I didn't read the previous discussion yet (will try tomorrow), sorry
> if I am off-topic. But I think the nastiest problem with tasklist
> is that it protects parent/child relationship. We need per-process
> lock, but first we should change ptrace to avoid this lock somehow.
> (this is one of the goals of ptrace-utrace, but not "immediate").

Didn't Thomas and you also come up with a scheme to push most signal
processing into task context?

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/