[patch 7/7] sched: only use TTWU_QUEUE when waker/wakee CPUs do not share top level cache

[Mike Galbraith <mgalbraith@suse.de>]

TTWU_QUEUE IPI overhead was measured to be as much as 13% of netperf TCP_RR
overhead when waking to a shared cache.  Don't IPI unless we're waking cross
cache, where it can be a winner.

Suggested by Peterz, tested by /me.

Signed-off-by: Mike Galbraith <efault@gmx.de>

---
 kernel/sched/core.c |   24 +++++++++++++++++++++++-
 1 file changed, 23 insertions(+), 1 deletion(-)

Index: linux-3.0-tip/kernel/sched/core.c
===================================================================
--- linux-3.0-tip.orig/kernel/sched/core.c
+++ linux-3.0-tip/kernel/sched/core.c
@@ -1479,12 +1479,34 @@ static int ttwu_activate_remote(struct t
 #endif /* __ARCH_WANT_INTERRUPTS_ON_CTXSW */
 #endif /* CONFIG_SMP */
 
+static int ttwu_share_cache(int this_cpu, int cpu)
+{
+#ifndef CONFIG_X86
+	struct sched_domain *sd;
+	int ret = 0;
+
+	rcu_read_lock();
+	for_each_domain(this_cpu, sd) {
+		if (!cpumask_test_cpu(cpu, sched_domain_span(sd)))
+			continue;
+
+		ret = (sd->flags & SD_SHARE_PKG_RESOURCES);
+		break;
+	}
+	rcu_read_unlock();
+
+	return ret;
+#else
+	return per_cpu(cpu_llc_id, this_cpu) == per_cpu(cpu_llc_id, cpu);
+#endif
+}
+
 static void ttwu_queue(struct task_struct *p, int cpu)
 {
 	struct rq *rq = cpu_rq(cpu);
 
 #if defined(CONFIG_SMP)
-	if (sched_feat(TTWU_QUEUE) && cpu != smp_processor_id()) {
+	if (sched_feat(TTWU_QUEUE) && !ttwu_share_cache(smp_processor_id(), cpu)) {
 		sched_clock_cpu(cpu); /* sync clocks x-cpu */
 		ttwu_queue_remote(p, cpu);
 		return;