sched: Use jump labels to reduce overhead when bandwidth control is inactive

Now that the linkage of jump-labels has been fixed they show a measurable
improvement in overhead for the enabled-but-unused case.

Workload is:

 'taskset -c 0 perf stat --repeat 50 -e instructions,cycles,branches
  bash -c "for ((i=0;i<5;i++)); do $(dirname $0)/pipe-test 20000; done"'

There's a speedup for all situations:

                 instructions            cycles                  branches
 -------------------------------------------------------------------------
 Intel Westmere
 base            806611770               745895590               146765378
 +jumplabel      803090165 (-0.44%)      713381840 (-4.36%)      144561130

 AMD Barcelona
 base            824657415               740055589               148855354
 +jumplabel      821056910 (-0.44%)      737558389 (-0.34%)      146635229

Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20111108042736.560831357@google.com
Signed-off-by: Ingo Molnar <mingo@elte.hu>

1 files changed, 12 insertions, 3 deletions

diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index a78ed27..a608593 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1421,7 +1421,7 @@ static void __account_cfs_rq_runtime(struct cfs_rq *cfs_rq,
 static __always_inline void account_cfs_rq_runtime(struct cfs_rq *cfs_rq,
 						   unsigned long delta_exec)
 {
-	if (!cfs_rq->runtime_enabled)
+	if (!cfs_bandwidth_used() || !cfs_rq->runtime_enabled)
 		return;
 
 	__account_cfs_rq_runtime(cfs_rq, delta_exec);
@@ -1429,13 +1429,13 @@ static __always_inline void account_cfs_rq_runtime(struct cfs_rq *cfs_rq,
 
 static inline int cfs_rq_throttled(struct cfs_rq *cfs_rq)
 {
-	return cfs_rq->throttled;
+	return cfs_bandwidth_used() && cfs_rq->throttled;
 }
 
 /* check whether cfs_rq, or any parent, is throttled */
 static inline int throttled_hierarchy(struct cfs_rq *cfs_rq)
 {
-	return cfs_rq->throttle_count;
+	return cfs_bandwidth_used() && cfs_rq->throttle_count;
 }
 
 /*
@@ -1756,6 +1756,9 @@ static void __return_cfs_rq_runtime(struct cfs_rq *cfs_rq)
 
 static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq)
 {
+	if (!cfs_bandwidth_used())
+		return;
+
 	if (!cfs_rq->runtime_enabled || cfs_rq->nr_running)
 		return;
 
@@ -1801,6 +1804,9 @@ static void do_sched_cfs_slack_timer(struct cfs_bandwidth *cfs_b)
  */
 static void check_enqueue_throttle(struct cfs_rq *cfs_rq)
 {
+	if (!cfs_bandwidth_used())
+		return;
+
 	/* an active group must be handled by the update_curr()->put() path */
 	if (!cfs_rq->runtime_enabled || cfs_rq->curr)
 		return;
@@ -1818,6 +1824,9 @@ static void check_enqueue_throttle(struct cfs_rq *cfs_rq)
 /* conditionally throttle active cfs_rq's from put_prev_entity() */
 static void check_cfs_rq_runtime(struct cfs_rq *cfs_rq)
 {
+	if (!cfs_bandwidth_used())
+		return;
+
 	if (likely(!cfs_rq->runtime_enabled || cfs_rq->runtime_remaining > 0))
 		return;