sched/core: Make sched_ttwu_pending() atomic in time (77558e4d) · Commits · e / devices / android_kernel_oneplus_sm7250

Since all tasks on the wake_list are woken under a single rq->lock avoid calling update_rq_clock() for each task. Signed-off-by:

Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by:

Ingo Molnar <mingo@kernel.org>

kernel/sched/core.c

+3 −1

Original line number	Diff line number	Diff line
		@@ -1685,7 +1685,7 @@ static void
		ttwu_do_activate(struct rq rq, struct task_struct p, int wake_flags,
		struct rq_flags *rf)
		{
		int en_flags = ENQUEUE_WAKEUP;
		int en_flags = ENQUEUE_WAKEUP \| ENQUEUE_NOCLOCK;

		lockdep_assert_held(&rq->lock);

		@@ -1737,6 +1737,7 @@ void sched_ttwu_pending(void)
		return;

		rq_lock_irqsave(rq, &rf);
		update_rq_clock(rq);

		while (llist) {
		int wake_flags = 0;
		@@ -1849,6 +1850,7 @@ static void ttwu_queue(struct task_struct *p, int cpu, int wake_flags)
		#endif

		rq_lock(rq, &rf);
		update_rq_clock(rq);
		ttwu_do_activate(rq, p, wake_flags, &rf);
		rq_unlock(rq, &rf);
		}