On Wed, 28 Aug 2019, Ming Lei wrote: > On Tue, Aug 27, 2019 at 06:19:00PM +0200, Thomas Gleixner wrote: > > > 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. > > > > That said, we already have infrastructure for something like this in the > > core interrupt code which is optimized to be lightweight in the fast path. > > > > kernel/irq/timings.c > > irq/timings.c is much more complicated, especially it applies EWMA to > compute each irq's average interval. However, we only need it for > do_IRQ() to cover all interrupt & softirq handler. That does not justify yet another ad hoc thingy. > Also CONFIG_IRQ_TIMINGS is usually disabled on distributions. That's not an argument at all. Thanks, tglx
