From: Suresh Siddha <suresh.b.siddha@intel.com>
To: Peter Zijlstra <peterz@infradead.org>,
Ingo Molnar <mingo@elte.hu>, Thomas Gleixner <tglx@linutronix.de>,
Arjan van de Ven <arjan@linux.jf.intel.com>
Cc: Venkatesh Pallipadi <venki@google.com>,
Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>,
ego@in.ibm.com, LKML <linux-kernel@vger.kernel.org>,
Dominik Brodowski <linux@dominikbrodowski.net>,
Nigel Cunningham <ncunningham@crca.org.au>,
Suresh Siddha <suresh.b.siddha@intel.com>
Subject: [patch 7/7] timers: use nearest busy cpu for migrating timers from an idle cpu
Date: Mon, 17 May 2010 11:27:33 -0700 [thread overview]
Message-ID: <20100517184028.114595207@sbs-t61.sc.intel.com> (raw)
In-Reply-To: <20100517182726.089700767@sbs-t61.sc.intel.com>
[-- Attachment #1: use_nearest_busy_cpu_for_migrating_timers_from_idle_cpu.patch --]
[-- Type: text/plain, Size: 3428 bytes --]
Currently we are migrating the unpinned timers from an idle to the cpu
doing idle load balancing (when all the cpus in the system are idle,
there is no idle load balacncing cpu and timers get added to the same idle cpu
where the request was made. So the current optimization works only on semi idle
system).
And In semi idle system, we no longer have periodic ticks on the idle cpu
doing the idle load balancing on behalf of all the cpu's. Using that cpu
will add more delays to the timers than intended (as that cpu's timer base
may not be uptodate wrt jiffies etc). This was causing mysterious slowdowns
during boot etc.
For now, in the semi idle case, use the nearest busy cpu for migrating timers from an
idle cpu. This is good for power-savings anyway.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
---
include/linux/sched.h | 2 +-
kernel/hrtimer.c | 8 ++------
kernel/sched.c | 13 +++++++++++++
kernel/timer.c | 8 ++------
4 files changed, 18 insertions(+), 13 deletions(-)
Index: tip/kernel/hrtimer.c
===================================================================
--- tip.orig/kernel/hrtimer.c
+++ tip/kernel/hrtimer.c
@@ -144,12 +144,8 @@ struct hrtimer_clock_base *lock_hrtimer_
static int hrtimer_get_target(int this_cpu, int pinned)
{
#ifdef CONFIG_NO_HZ
- if (!pinned && get_sysctl_timer_migration() && idle_cpu(this_cpu)) {
- int preferred_cpu = get_nohz_load_balancer();
-
- if (preferred_cpu < nr_cpu_ids)
- return preferred_cpu;
- }
+ if (!pinned && get_sysctl_timer_migration() && idle_cpu(this_cpu))
+ return get_nohz_timer_target();
#endif
return this_cpu;
}
Index: tip/kernel/sched.c
===================================================================
--- tip.orig/kernel/sched.c
+++ tip/kernel/sched.c
@@ -1201,6 +1201,19 @@ static void resched_cpu(int cpu)
}
#ifdef CONFIG_NO_HZ
+int get_nohz_timer_target(void)
+{
+ int cpu = smp_processor_id();
+ int i;
+ struct sched_domain *sd;
+
+ for_each_domain(cpu, sd) {
+ for_each_cpu(i, sched_domain_span(sd))
+ if (!idle_cpu(i))
+ return i;
+ }
+ return cpu;
+}
/*
* When add_timer_on() enqueues a timer into the timer wheel of an
* idle CPU then this timer might expire before the next timer event
Index: tip/kernel/timer.c
===================================================================
--- tip.orig/kernel/timer.c
+++ tip/kernel/timer.c
@@ -679,12 +679,8 @@ __mod_timer(struct timer_list *timer, un
cpu = smp_processor_id();
#if defined(CONFIG_NO_HZ) && defined(CONFIG_SMP)
- if (!pinned && get_sysctl_timer_migration() && idle_cpu(cpu)) {
- int preferred_cpu = get_nohz_load_balancer();
-
- if (preferred_cpu < nr_cpu_ids)
- cpu = preferred_cpu;
- }
+ if (!pinned && get_sysctl_timer_migration() && idle_cpu(cpu))
+ cpu = get_nohz_timer_target();
#endif
new_base = per_cpu(tvec_bases, cpu);
Index: tip/include/linux/sched.h
===================================================================
--- tip.orig/include/linux/sched.h
+++ tip/include/linux/sched.h
@@ -273,7 +273,7 @@ extern void task_rq_unlock_wait(struct t
extern cpumask_var_t nohz_cpu_mask;
#if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ)
extern void select_nohz_load_balancer(int stop_tick);
-extern int get_nohz_load_balancer(void);
+extern int get_nohz_timer_target(void);
extern int nohz_ratelimit(int cpu);
#else
static inline void select_nohz_load_balancer(int stop_tick) { }
next prev parent reply other threads:[~2010-05-17 19:02 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-05-17 18:27 [patch 0/7] sched: change nohz idle load balancing logic to push model Suresh Siddha
2010-05-17 18:27 ` [patch 1/7] softirq: Add a no local fallback option to send_remote_softirq Suresh Siddha
2010-05-17 18:27 ` [patch 2/7] softirq: add init_remote_softirq_csd() Suresh Siddha
2010-05-17 18:27 ` [patch 3/7] softirq: avoid softirq_work_list for SCHED_SOFTIRQ when sent remotely Suresh Siddha
2010-05-20 8:12 ` Peter Zijlstra
2010-05-20 8:14 ` David Miller
2010-05-20 8:23 ` Jens Axboe
2010-05-20 8:29 ` Peter Zijlstra
2010-05-20 9:18 ` David Miller
2010-05-17 18:27 ` [patch 4/7] sched: Change nohz ilb logic from pull to push model Suresh Siddha
2010-06-01 23:47 ` Vaidyanathan Srinivasan
2010-06-02 22:27 ` Suresh Siddha
2010-05-17 18:27 ` [patch 5/7] sched: Change select_nohz_load_balancer to return void Suresh Siddha
2010-05-17 18:27 ` [patch 6/7] sched: change nohz.load_balancer to be nr_cpu_ids based Suresh Siddha
2010-05-20 9:49 ` Peter Zijlstra
2010-05-17 18:27 ` Suresh Siddha [this message]
2010-06-01 23:37 ` [patch 7/7] timers: use nearest busy cpu for migrating timers from an idle cpu Vaidyanathan Srinivasan
2010-06-02 22:02 ` Suresh Siddha
2010-05-17 22:39 ` [patch 0/7] sched: change nohz idle load balancing logic to push model Nigel Cunningham
2010-05-19 9:19 ` Dominik Brodowski
2010-05-20 10:50 ` Peter Zijlstra
2010-05-22 0:09 ` Suresh Siddha
2010-05-31 9:17 ` Peter Zijlstra
2010-06-09 10:13 ` [tip:sched/core] sched: Change " tip-bot for Venkatesh Pallipadi
2010-05-20 11:07 ` [patch 0/7] sched: change " Nigel Cunningham
2010-05-20 11:17 ` Dominik Brodowski
2010-05-20 11:35 ` Nigel Cunningham
2010-05-20 12:13 ` Dominik Brodowski
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=20100517184028.114595207@sbs-t61.sc.intel.com \
--to=suresh.b.siddha@intel.com \
--cc=arjan@linux.jf.intel.com \
--cc=ego@in.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux@dominikbrodowski.net \
--cc=mingo@elte.hu \
--cc=ncunningham@crca.org.au \
--cc=peterz@infradead.org \
--cc=svaidy@linux.vnet.ibm.com \
--cc=tglx@linutronix.de \
--cc=venki@google.com \
/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
all inboxes | Powered by JetHome®