Re: [PATCH] cpu/hotplug: Serialize callback invocations proper

From: Thomas Gleixner
Date: Tue Mar 14 2017 - 13:44:07 EST


On Tue, 14 Mar 2017, Bart Van Assche wrote:
> On Tue, 2017-03-14 at 16:06 +0100, Sebastian Andrzej Siewior wrote:
> > The setup/remove_state/instance() functions in the hotplug core code are
> > serialized against concurrent CPU hotplug, but unfortunately not serialized
> > against themself.
> >
> > As a consequence a concurrent invocation of these function results in
> > corruption of the callback machinery because two instances try to invoke
> > callbacks on remote cpus at the same time. This results in missing callback
> > invocations and initiator threads waiting forever on the completion.
> >
> > The obvious solution to replace get_cpu_online() with cpu_hotplug_begin()
> > is not possible because at least one callsite calls into these functions
> > from a get_online_cpu() locked region.
> >
> > Extend the protection scope of the cpuhp_state_mutex from solely protecting
> > the state arrays to cover the callback invocation machinery as well.
> >
> > Reported-by: Bart Van Assche <Bart.VanAssche@xxxxxxxxxxx>
> > Fixes: 5b7aa87e0482 ("cpu/hotplug: Implement setup/removal interface")
> > Signed-off-by: Sebastian Andrzej Siewior <bigeasy@xxxxxxxxxxxxx>
>
> Tested-by: Bart Van Assche <Bart.VanAssche@xxxxxxxxxxx>
>
> So this regression was introduced in kernel v4.6? Anyway, thanks for the patch!

And it's very timing sensitive ....

Thanks for trying to bisect this!

tglx