On Thu, Mar 17, 2016 at 12:36 PM, Peter Zijlstra [off-list ref] wrote:

On Thu, Mar 17, 2016 at 01:01:45AM +0100, Rafael J. Wysocki wrote:

quoted

+     } else if (sg_policy->next_freq != next_freq) {
+             sg_policy->work_cpu = smp_processor_id();
+             irq_work_queue(&sg_policy->irq_work);
+     }
+}

quoted

+static void sugov_irq_work(struct irq_work *irq_work)
+{
+     struct sugov_policy *sg_policy;
+
+     sg_policy = container_of(irq_work, struct sugov_policy, irq_work);
+     schedule_work_on(sg_policy->work_cpu, &sg_policy->work);
+}

Not sure I see the point of ->work_cpu, irq_work_queue() does guarantee
the same CPU, so the above is identical to:

        schedule_work_on(smp_processor_id(), &sq_policy->work);

OK

I'll do that and restore work_in_progress, then.