On 7/26/22 18:26, Steven Rostedt wrote:
> On Mon, 25 Jul 2022 22:11:14 +0200
> Daniel Bristot de Oliveira <bristot@xxxxxxxxxx> wrote:
>
>> +/**
>> + * reactor_cleanup_monitor - cleanup a monitor reference
>> + * @mdef: monitor's definition.
>> + */
>> +void reactor_cleanup_monitor(struct rv_monitor_def *mdef)
>> +{
>> + mdef->rdef->counter--;
>> +}
>> +
>
> Doesn't this require a lock? Also you might want to add a WARN_ONCE() if it
> goes below zero.
reactor_populate_monitor() and reactor_cleanup_monitor() should be called with
lock taken... adding lockdep check.
oops, I was not actually calling it yet. Not a BUG, as we do not unload reactors,
it was BUG on the way to happen :-)
Adding a WARN_ONCE().
-- Daniel
>
> -- Steve
