Re: [PATCH 1/2] sched_ext: Enable tick for finite slices on nohz_full

[Tejun Heo]

Hello, Andrea.

On Tue, Jul 07, 2026 at 10:17:59AM +0200, Andrea Righi wrote:
> 	if ((p->scx.slice == SCX_SLICE_INF) !=
> 	    (bool)(rq->scx.flags & SCX_RQ_CAN_STOP_TICK)) {
> 		if (p->scx.slice == SCX_SLICE_INF) {

The slice type is tested three times here (the XOR, the inner if, and the
finite tail). Can you split on it once instead?

	if (p->scx.slice == SCX_SLICE_INF) {
		if (!(rq->scx.flags & SCX_RQ_CAN_STOP_TICK)) {
			rq->scx.flags |= SCX_RQ_CAN_STOP_TICK;
			sched_update_tick_dependency(rq);
			update_other_load_avgs(rq);
		}
	} else {
		if (rq->scx.flags & SCX_RQ_CAN_STOP_TICK) {
			rq->scx.flags &= ~SCX_RQ_CAN_STOP_TICK;
			update_other_load_avgs(rq);
		}
		if (tick_nohz_full_cpu(cpu_of(rq)))
			tick_nohz_dep_set_cpu(cpu_of(rq), TICK_DEP_BIT_SCHED);
	}

update_other_load_avgs() lands in both arms, but the infinite/finite split
reads more directly than keying off the transition.

Thanks.

-- 
tejun