On Tue, Aug 27, 2019 at 04:42:02PM +0200, Thomas Gleixner wrote:
> On Tue, 27 Aug 2019, Ming Lei wrote:
> > +/*
> > + * Update average irq interval with the Exponential Weighted Moving
> > + * Average(EWMA)
> > + */
> > +static void irq_update_interval(void)
> > +{
> > +#define IRQ_INTERVAL_EWMA_WEIGHT 128
> > +#define IRQ_INTERVAL_EWMA_PREV_FACTOR 127
> > +#define IRQ_INTERVAL_EWMA_CURR_FACTOR (IRQ_INTERVAL_EWMA_WEIGHT - \
> > + IRQ_INTERVAL_EWMA_PREV_FACTOR)
>
> Please do not stick defines into a function body. That's horrible.
OK.
>
> > +
> > + int cpu = raw_smp_processor_id();
> > + struct irq_interval *inter = per_cpu_ptr(&avg_irq_interval, cpu);
> > + u64 delta = sched_clock_cpu(cpu) - inter->last_irq_end;
>
> Why are you doing that raw_smp_processor_id() dance? The call site has
> interrupts and preemption disabled.
OK, will change to __smp_processor_id().
>
> Also how is that supposed to work when sched_clock is jiffies based?
Good catch, looks ktime_get_ns() is needed.
>
> > + inter->avg = (inter->avg * IRQ_INTERVAL_EWMA_PREV_FACTOR +
> > + delta * IRQ_INTERVAL_EWMA_CURR_FACTOR) /
> > + IRQ_INTERVAL_EWMA_WEIGHT;
>
> We definitely are not going to have a 64bit multiplication and division on
> every interrupt. Asided of that this breaks 32bit builds all over the place.
I will convert the above into add/subtract/shift only.
thanks,
Ming
