linux-kernel - [PATCH 1/4] sched/fair: Introduce sched_entity::dont

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]

Date:	Mon, 20 Jun 2016 14:15:11 +0200
From:	Jiri Olsa <jolsa@...nel.org>
To:	Ingo Molnar <mingo@...nel.org>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>
Cc:	lkml <linux-kernel@...r.kernel.org>,
	James Hartsock <hartsjc@...hat.com>,
	Rik van Riel <riel@...hat.com>,
	Srivatsa Vaddagiri <vatsa@...ux.vnet.ibm.com>,
	Kirill Tkhai <ktkhai@...allels.com>
Subject: [PATCH 1/4] sched/fair: Introduce sched_entity::dont_balance

Adding dont_balance bool into struct sched_entity,
to mark tasks which are rebalanced based on affinity.

It's used only when REBALANCE_AFFINITY feature is
switched on. The code functionality of this feature
is introduced in following patch.

Signed-off-by: Jiri Olsa <jolsa@...nel.org>
---
 include/linux/sched.h   |  2 ++
 kernel/sched/fair.c     | 21 ++++++++++++++++++---
 kernel/sched/features.h |  1 +
 3 files changed, 21 insertions(+), 3 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index dee41bf59e6b..0e6ac882283b 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1331,6 +1331,8 @@ struct sched_entity {
 
 	u64			nr_migrations;
 
+	bool			dont_balance;
+
 #ifdef CONFIG_SCHEDSTATS
 	struct sched_statistics statistics;
 #endif
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index c6dd8bab010c..f19c9435c64d 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -900,6 +900,11 @@ update_stats_curr_start(struct cfs_rq *cfs_rq, struct sched_entity *se)
 	se->exec_start = rq_clock_task(rq_of(cfs_rq));
 }
 
+static bool dont_balance(struct task_struct *p)
+{
+	return sched_feat(REBALANCE_AFFINITY) && p->se.dont_balance;
+}
+
 /**************************************************
  * Scheduling class queueing methods:
  */
@@ -2172,6 +2177,10 @@ void task_numa_fault(int last_cpupid, int mem_node, int pages, int flags)
 	if (!static_branch_likely(&sched_numa_balancing))
 		return;
 
+	/* Don't move task if the affinity balance is active. */
+	if (dont_balance(p))
+		return;
+
 	/* for example, ksmd faulting in a user's mm */
 	if (!p->mm)
 		return;
@@ -3387,6 +3396,8 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags)
 
 	update_min_vruntime(cfs_rq);
 	update_cfs_shares(cfs_rq);
+
+	se->dont_balance = false;
 }
 
 /*
@@ -6017,13 +6028,17 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env)
 	/*
 	 * We do not migrate tasks that are:
 	 * 1) throttled_lb_pair, or
-	 * 2) cannot be migrated to this CPU due to cpus_allowed, or
-	 * 3) running (obviously), or
-	 * 4) are cache-hot on their current CPU.
+	 * 2) dont_balance is set, or
+	 * 3) cannot be migrated to this CPU due to cpus_allowed, or
+	 * 4) running (obviously), or
+	 * 5) are cache-hot on their current CPU.
 	 */
 	if (throttled_lb_pair(task_group(p), env->src_cpu, env->dst_cpu))
 		return 0;
 
+	if (dont_balance(p))
+		return 0;
+
 	if (!cpumask_test_cpu(env->dst_cpu, tsk_cpus_allowed(p))) {
 		int cpu;
 
diff --git a/kernel/sched/features.h b/kernel/sched/features.h
index 69631fa46c2f..acdd78a6be4d 100644
--- a/kernel/sched/features.h
+++ b/kernel/sched/features.h
@@ -68,4 +68,5 @@ SCHED_FEAT(FORCE_SD_OVERLAP, false)
 SCHED_FEAT(RT_RUNTIME_SHARE, true)
 SCHED_FEAT(LB_MIN, false)
 SCHED_FEAT(ATTACH_AGE_LOAD, true)
+SCHED_FEAT(REBALANCE_AFFINITY, false)
 
-- 
2.4.11