linux-kernel - [patch 4/6] sched: ratelimit select_idle

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: <1321629659.7080.19.camel@marge.simson.net>
Date:	Fri, 18 Nov 2011 16:20:59 +0100
From:	Mike Galbraith <efault@....de>
To:	Suresh Siddha <suresh.b.siddha@...el.com>
Cc:	Peter Zijlstra <peterz@...radead.org>,
	linux-kernel <linux-kernel@...r.kernel.org>,
	Ingo Molnar <mingo@...e.hu>, Paul Turner <pjt@...gle.com>
Subject: [patch 4/6] sched: ratelimit select_idle_sibling()for sync wakeups


select_idle_sibling() injures synchromous loads horribly
on processors with L3 (ala westmere) when called at high
Frequency.  Cut it off at 40 KHz (25usec event rate) when
waking sync, in lieu of an inter-domain cache penalty.

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

---
 kernel/sched_fair.c     |   20 ++++++++++++++++++--
 kernel/sched_features.h |    6 ++++++
 2 files changed, 24 insertions(+), 2 deletions(-)

Index: linux-3.2.git/kernel/sched_fair.c
===================================================================
--- linux-3.2.git.orig/kernel/sched_fair.c
+++ linux-3.2.git/kernel/sched_fair.c
@@ -2264,6 +2264,21 @@ static inline struct sched_domain *highe
 }
 
 /*
+ * select_idle_sibling() injures synchromous loads horribly
+ * on processors with L3 (ala westmere) when called at high
+ * Frequency.  Cut it off at 40 KHz (25usec event rate) when
+ * waking sync, in lieu of an inter-domain cache penalty.
+ */
+#define SIBLING_SYNC_CUTOFF_NS (NSEC_PER_SEC/40000UL)
+
+static int idle_sibling_limit(int target, int sync)
+{
+	if (!sync || !sched_feat(SIBLING_LIMIT_SYNC))
+		return 0;
+	return cpu_rq(target)->avg_event < SIBLING_SYNC_CUTOFF_NS;
+}
+
+/*
  * Try and locate an idle CPU in the sched_domain.
  */
 static int select_idle_sibling(struct task_struct *p, int target)
@@ -2400,9 +2415,10 @@ select_task_rq_fair(struct task_struct *
 
 	if (affine_sd) {
 		if (cpu == prev_cpu || wake_affine(affine_sd, p, sync))
-			prev_cpu = cpu;
+			new_cpu = cpu;
 
-		new_cpu = select_idle_sibling(p, prev_cpu);
+		if (!idle_sibling_limit(new_cpu, sync))
+			new_cpu = select_idle_sibling(p, new_cpu);
 		goto unlock;
 	}
 
Index: linux-3.2.git/kernel/sched_features.h
===================================================================
--- linux-3.2.git.orig/kernel/sched_features.h
+++ linux-3.2.git/kernel/sched_features.h
@@ -67,3 +67,9 @@ SCHED_FEAT(NONTASK_POWER, 1)
 SCHED_FEAT(TTWU_QUEUE, 1)
 
 SCHED_FEAT(FORCE_SD_OVERLAP, 0)
+
+/*
+ * Restrict the frequency at which select_idle_sibling() may be called
+ * for synchronous wakeups.
+ */
+SCHED_FEAT(SIBLING_LIMIT_SYNC, 1)


--
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/