public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Mike Galbraith <efault@gmx.de>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Ingo Molnar <mingo@elte.hu>, LKML <linux-kernel@vger.kernel.org>
Subject: Re: [patch 1/12] sched: ratelimit nohz
Date: Thu, 11 Mar 2010 10:50:03 +0100	[thread overview]
Message-ID: <1268301003.6785.28.camel@marge.simson.net> (raw)
In-Reply-To: <1268300950.6785.27.camel@marge.simson.net>


sched: ratelimit nohz

Entering nohz code on every micro-idle is costing ~10% throughput for netperf
TCP_RR when scheduling cross-cpu.  Rate limiting entry fixes this, but raises
ticks a bit.  On my Q6600, an idle box goes from ~85 interrupts/sec to 128.

The higher the context switch rate, the more nohz entry costs.  With this patch
and some cycle recovery patches in my tree, max cross cpu context switch rate is
improved by ~16%, a large portion of which of which is this ratelimiting.

Signed-off-by: Mike Galbraith <efault@gmx.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>

---
 include/linux/sched.h    |    6 ++++++
 kernel/sched.c           |   12 ++++++++++++
 kernel/time/tick-sched.c |    3 +++
 3 files changed, 21 insertions(+)

Index: linux-2.6/include/linux/sched.h
===================================================================
--- linux-2.6.orig/include/linux/sched.h
+++ linux-2.6/include/linux/sched.h
@@ -271,11 +271,17 @@ extern cpumask_var_t nohz_cpu_mask;
 #if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ)
 extern int select_nohz_load_balancer(int cpu);
 extern int get_nohz_load_balancer(void);
+extern int nohz_ratelimit(int cpu);
 #else
 static inline int select_nohz_load_balancer(int cpu)
 {
 	return 0;
 }
+
+static inline int nohz_ratelimit(int cpu)
+{
+	return 0;
+}
 #endif
 
 /*
Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -492,6 +492,7 @@ struct rq {
 	#define CPU_LOAD_IDX_MAX 5
 	unsigned long cpu_load[CPU_LOAD_IDX_MAX];
 #ifdef CONFIG_NO_HZ
+	u64 nohz_stamp;
 	unsigned char in_nohz_recently;
 #endif
 	/* capture load from *all* tasks on this cpu: */
@@ -1228,6 +1229,17 @@ void wake_up_idle_cpu(int cpu)
 	if (!tsk_is_polling(rq->idle))
 		smp_send_reschedule(cpu);
 }
+
+int nohz_ratelimit(int cpu)
+{
+	struct rq *rq = cpu_rq(cpu);
+	u64 diff = rq->clock - rq->nohz_stamp;
+
+	rq->nohz_stamp = rq->clock;
+
+	return diff < (NSEC_PER_SEC / HZ) >> 1;
+}
+
 #endif /* CONFIG_NO_HZ */
 
 static u64 sched_avg_period(void)
Index: linux-2.6/kernel/time/tick-sched.c
===================================================================
--- linux-2.6.orig/kernel/time/tick-sched.c
+++ linux-2.6/kernel/time/tick-sched.c
@@ -262,6 +262,9 @@ void tick_nohz_stop_sched_tick(int inidl
 		goto end;
 	}
 
+	if (nohz_ratelimit(cpu))
+		goto end;
+
 	ts->idle_calls++;
 	/* Read jiffies and the time when jiffies were updated last */
 	do {



  reply	other threads:[~2010-03-11  9:50 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-03-11  9:49 [patch 0/12] sched: fastpath cycle recovery Mike Galbraith
2010-03-11  9:50 ` Mike Galbraith [this message]
2010-03-11 18:30   ` [tip:sched/core] sched: Rate-limit nohz tip-bot for Mike Galbraith
2010-03-11  9:51 ` [patch 2/12] sched: remove avg_wakeup Mike Galbraith
2010-03-11 18:30   ` [tip:sched/core] sched: Remove avg_wakeup tip-bot for Mike Galbraith
2010-03-11  9:52 ` [patch 3/12] sched: remove avg_overlap Mike Galbraith
2010-03-11 18:31   ` [tip:sched/core] sched: Remove avg_overlap tip-bot for Mike Galbraith
2010-03-11  9:53 ` [patch 4/12] sched: cleanup/optimize clock updates Mike Galbraith
2010-03-11 18:31   ` [tip:sched/core] sched: Cleanup/optimize " tip-bot for Mike Galbraith
2010-03-11  9:54 ` [patch 5/12] sched: tweak sched_latency and min_granularity Mike Galbraith
2010-03-11 18:31   ` [tip:sched/core] sched: Tweak " tip-bot for Mike Galbraith
2010-03-11  9:56 ` [patch 6/12] sched: fix select_idle_sibling() Mike Galbraith
2010-03-11 18:32   ` [tip:sched/core] sched: Fix select_idle_sibling() tip-bot for Mike Galbraith
2010-03-11  9:57 ` [patch 7/12] sched: remove NORMALIZED_SLEEPER Mike Galbraith
2010-03-11 18:32   ` [tip:sched/core] sched: Remove NORMALIZED_SLEEPER tip-bot for Mike Galbraith
2010-03-11  9:58 ` [patch 8/12] sched: remove FAIR_SLEEPERS feature Mike Galbraith
2010-03-11 18:32   ` [tip:sched/core] sched: Remove " tip-bot for Mike Galbraith
2010-03-11  9:59 ` [patch 9/12] sched: remove WAKEUP_SYNC feature Mike Galbraith
2010-03-11 18:32   ` [tip:sched/core] sched: Remove " tip-bot for Mike Galbraith
2010-03-11 10:01 ` [patch 11/12] sched: remove ASYM_GRAN feature Mike Galbraith
2010-03-11 18:33   ` [tip:sched/core] sched: Remove " tip-bot for Mike Galbraith
2010-03-11 10:03 ` [patch 10/12] sched: remove SYNC_WAKEUPS feature Mike Galbraith
2010-03-11 18:33   ` [tip:sched/core] sched: Remove " tip-bot for Mike Galbraith
2010-03-11 10:04 ` [patch 12/12] sched: remove AFFINE_WAKEUPS feature Mike Galbraith
2010-03-11 18:33   ` [tip:sched/core] sched: Remove " tip-bot for Mike Galbraith
2010-03-12  3:23     ` Yong Zhang
2010-03-12  4:37       ` Mike Galbraith

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1268301003.6785.28.camel@marge.simson.net \
    --to=efault@gmx.de \
    --cc=a.p.zijlstra@chello.nl \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox