Messages in this thread | | | Date | Thu, 10 Aug 2017 16:23:52 +0200 | From | Ingo Molnar <> | Subject | Re: [PATCH RESEND] x86/smpboot: Unbreak CPU0 hotplug |
| |
* Vitaly Kuznetsov <vkuznets@redhat.com> wrote:
> Vitaly Kuznetsov <vkuznets@redhat.com> writes: > > > A hang on CPU0 onlining after a preceding offlining is observed. Trace > > shows that CPU0 is stuck in check_tsc_sync_target() waiting for source > > CPU to run check_tsc_sync_source() but this never happens. Source CPU, > > in its turn, is stuck on synchronize_sched() which is called from > > native_cpu_up() -> do_boot_cpu() -> unregister_nmi_handler(). > > > > Fix the issue by moving unregister_nmi_handler() from do_boot_cpu() to > > native_cpu_up() after cpu onlining is done.
Looks like a classic ABBA deadlock, due to the use of synchronize_sched() in unregister_nmi_handler(), right?
> > > > Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com> > > --- > > It's been awile since my v1 submission, no comments so far. Resending. > > Sorry, but > > ping? > > I haven't received a single comment on this since the initial submission > on June, 26 - is it so bad? :-)
So the fix looks good to me at first sight, but wanted to wait for Thomas to ack it - once he gets back from vacation.
Thanks,
Ingo
| |