linux-kernel - [RFC PATCH 13/13] sched: Modifying wake

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20121025102622.21022.59828.stgit@preeti.in.ibm.com>
Date:	Thu, 25 Oct 2012 15:56:22 +0530
From:	Preeti U Murthy <preeti@...ux.vnet.ibm.com>
To:	svaidy@...ux.vnet.ibm.com, linux-kernel@...r.kernel.org
Cc:	mingo@...nel.org, venki@...gle.com, robin.randhawa@....com,
	linaro-dev@...ts.linaro.org, a.p.zijlstra@...llo.nl,
	mjg59@...f.ucam.org, viresh.kumar@...aro.org,
	akpm@...ux-foundation.org, amit.kucheria@...aro.org,
	deepthi@...ux.vnet.ibm.com, paul.mckenney@...aro.org,
	arjan@...ux.intel.com, paulmck@...ux.vnet.ibm.com,
	srivatsa.bhat@...ux.vnet.ibm.com, vincent.guittot@...aro.org,
	tglx@...utronix.de, Arvind.Chauhan@....com, pjt@...gle.com,
	Morten.Rasmussen@....com, linux-arm-kernel@...ts.infradead.org,
	suresh.b.siddha@...el.com
Subject: [RFC PATCH 13/13] sched: Modifying wake_affine to use PJT's metric

Additional parameters introduced to perform this function which are
calculated using PJT's metrics and its helpers.

Signed-off-by: Preeti U Murthy <preeti@...ux.vnet.ibm.com>
---
 kernel/sched/fair.c |   34 +++++++++++++++-------------------
 1 file changed, 15 insertions(+), 19 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 15ec528..b4b572c 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -2931,19 +2931,6 @@ static unsigned long power_of(int cpu)
 	return cpu_rq(cpu)->cpu_power;
 }
 
-static unsigned long cpu_avg_load_per_task(int cpu)
-{
-	struct rq *rq = cpu_rq(cpu);
-	unsigned long nr_running = ACCESS_ONCE(rq->nr_running);
-
-	if (nr_running) {
-		return rq->load.weight / nr_running;
-	}
-
-	return 0;
-}
-
-
 static void task_waking_fair(struct task_struct *p)
 {
 	struct sched_entity *se = &p->se;
@@ -3085,16 +3072,18 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync)
 {
 	s64 this_load, load;
 	int idx, this_cpu, prev_cpu;
-	unsigned long tl_per_task;
+	u64 tl_per_task; /* Modified to reflect PJT's metric */
 	struct task_group *tg;
-	unsigned long weight;
+	unsigned long weight, nr_running;
 	int balanced;
 
 	idx	  = sd->wake_idx;
 	this_cpu  = smp_processor_id();
 	prev_cpu  = task_cpu(p);
-	load	  = source_load(prev_cpu, idx);
-	this_load = target_load(this_cpu, idx);
+	/* Both of the below have been modified to use PJT's metric */
+	load      = cpu_rq(prev_cpu)->cfs.runnable_load_avg;
+	this_load = cpu_rq(this_cpu)->cfs.runnable_load_avg;
+	nr_running = cpu_rq(this_cpu)->nr_running;
 
 	/*
 	 * If sync wakeup then subtract the (maximum possible)
@@ -3104,6 +3093,7 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync)
 	if (sync) {
 		tg = task_group(current);
 		weight = current->se.load.weight;
+		weight = current->se.avg.load_avg_contrib;
 
 		this_load += effective_load(tg, this_cpu, -weight, -weight);
 		load += effective_load(tg, prev_cpu, 0, -weight);
@@ -3111,6 +3101,8 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync)
 
 	tg = task_group(p);
 	weight = p->se.load.weight;
+	/* The below change to reflect PJT's metric */
+	weight = p->se.avg.load_avg_contrib;
 
 	/*
 	 * In low-load situations, where prev_cpu is idle and this_cpu is idle
@@ -3146,11 +3138,15 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync)
 		return 1;
 
 	schedstat_inc(p, se.statistics.nr_wakeups_affine_attempts);
-	tl_per_task = cpu_avg_load_per_task(this_cpu);
+	/* Below modification to use PJT's metric */
+	if (nr_running)
+		tl_per_task = cpu_rq(this_cpu)->cfs.runnable_load_avg / nr_running;
+	else
+		tl_per_task = 0;
 
 	if (balanced ||
 	    (this_load <= load &&
-	     this_load + target_load(prev_cpu, idx) <= tl_per_task)) {
+	     this_load + cpu_rq(prev_cpu)->cfs.runnable_load_avg <= tl_per_task)) {
 		/*
 		 * This domain has SD_WAKE_AFFINE and
 		 * p is cache cold in this domain, and

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/