Messages in this thread | | | Date | Fri, 20 Oct 2017 20:14:24 +0200 (CEST) | From | Thomas Gleixner <> | Subject | Re: Possible bug in CPU hotplug handling (4.14.0-rc5) |
| |
On Fri, 20 Oct 2017, Tvrtko Ursulin wrote:
> Hi all, > > I think I've found in bug in the CPU hotplug handling when multi-instance > states are used. That is in the 4.14.0-rc5 kernel.
> I have not attempted to get to the bottom of the issue to propose an actual > fix, since the logic there looks somewhat complex, but thought to first > seek opinion of the people in the know regarding this area.
These commits fiddled with that post rc1, so it's probably burried somewhere there.
1db49484f21e ("smp/hotplug: Hotplug state fail injection") 5ebe7742fff8 ("smp/hotplug: Differentiate the AP completion between up and down") 5f4b55e10645 ("smp/hotplug: Differentiate the AP-work lockdep class between up and down") 724a86881d03 ("smp/hotplug: Callback vs state-machine consistency") 4dddfb5faa61 ("smp/hotplug: Rewrite AP state machine core") 96abb968549c ("smp/hotplug: Allow external multi-instance rollback")
I'll have a look at it on sunday (tomorrow I'm conferencing) with Peter, so we should be able to figure it out, unless you beat us to it.
Thanks,
tglx
| |