Re: Possible race between CPU hotplug and perf_pmu_migrate_context

[Mark Rutland]

Hi all,

Further to my earlier reply I've come up with a potential fix below,
which has survived my stress test for both my WIP driver and the intel
uncore imc driver.

As it's impossible to synchronize with the event->ctx I'd hoped it would
be possible to synchronize with a field on the event itself.
Unfortunately all I managed to come up with were some shiny new ABBA
deadlocks.

Instead I've followed the example set by perf_event_open and inhibited
CPU hotplug for the portion of put_event that removes an event from its
context, which will prevent perf_pmu_migrate_context from modifying
event->ctx under our feet.

While there's the potential to starve CPU hotplug, that's already the
case for the perf_event_open path, so I'm not sure if that's a big deal
or not.

Thoughts?

Mark.

---->8----