mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Arjan van de Ven <arjan@linux.intel.com>
To: linux-kernel@vger.kernel.org
Cc: mingo@elte.hu, akpm@osdl.org
Subject: [patch 1/2] round_jiffies infrastructure
Date: Tue, 10 Oct 2006 18:03:30 +0200	[thread overview]
Message-ID: <1160496210.3000.310.camel@laptopd505.fenrus.org> (raw)
In-Reply-To: <1160496165.3000.308.camel@laptopd505.fenrus.org>

From: Arjan van de Ven <arjan@linux.intel.com>
Subject: round_jiffies infrastructure

This patch introduces a round_jiffies() function as well as a
round_jiffies_relative() function. These functions round a jiffies value
to the next whole second. The primary purpose of this rounding is to
cause all "we don't care exactly when" timers to happen at the same jiffy.
This avoids multiple timers to fire within the second for no real reason;
with dynamic ticks these extra timers cause wakeups from deep sleep
CPU sleep states and thus waste power. 

The exact wakeup moment is skewed by the cpu number, to avoid all
cpus from waking up at the exact same time (and hitting the same 
lock/cachelines there)

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>



Index: linux-2.6.19-rc1-git6/include/linux/timer.h
===================================================================
--- linux-2.6.19-rc1-git6.orig/include/linux/timer.h
+++ linux-2.6.19-rc1-git6/include/linux/timer.h
@@ -98,4 +98,10 @@ extern void run_local_timers(void);
 struct hrtimer;
 extern int it_real_fn(struct hrtimer *);
 
+unsigned long __round_jiffies(unsigned long T, int CPU);
+unsigned long __round_jiffies_relative(unsigned long T, int CPU);
+unsigned long round_jiffies(unsigned long T);
+unsigned long round_jiffies_relative(unsigned long T);
+
+
 #endif
Index: linux-2.6.19-rc1-git6/kernel/timer.c
===================================================================
--- linux-2.6.19-rc1-git6.orig/kernel/timer.c
+++ linux-2.6.19-rc1-git6/kernel/timer.c
@@ -80,6 +80,56 @@ tvec_base_t boot_tvec_bases;
 EXPORT_SYMBOL(boot_tvec_bases);
 static DEFINE_PER_CPU(tvec_base_t *, tvec_bases) = &boot_tvec_bases;
 
+unsigned long __round_jiffies(unsigned long T, int CPU)
+{
+	int rem;
+	int original  = T;
+	rem = T % HZ;
+	if (rem < HZ/4)
+		T = T - rem;
+	else
+		T = T - rem + HZ;
+	/* we don't want all cpus firing at once hitting the same lock/memory */
+	T += CPU * 3;
+	if (T <= jiffies) /* rounding ate our timeout entirely */
+		return original;
+	return T;
+}
+EXPORT_SYMBOL_GPL(__round_jiffies);
+
+unsigned long __round_jiffies_relative(unsigned long T, int CPU)
+{
+	int rem;
+	int original = T;
+	T=T+jiffies;
+	rem = T % HZ;
+	if (rem < HZ/4)
+		T = T - rem;
+	else
+		T = T - rem + HZ;
+	/* we don't want all cpus firing at once hitting the same lock/memory */
+	T += CPU * 3;
+	T = T-jiffies;
+	if (T<=0) /* rounding ate our delay entirely, don't round */
+		return original;
+	return T;
+}
+EXPORT_SYMBOL_GPL(__round_jiffies_relative);
+
+unsigned long round_jiffies(unsigned long T)
+{
+	return __round_jiffies(T, raw_smp_processor_id());
+}
+EXPORT_SYMBOL_GPL(round_jiffies);
+
+unsigned long round_jiffies_relative(unsigned long T)
+{
+	return __round_jiffies_relative(T, raw_smp_processor_id());
+}
+EXPORT_SYMBOL_GPL(round_jiffies_relative);
+
+
+
 static inline void set_running_timer(tvec_base_t *base,
 					struct timer_list *timer)
 {


  reply	other threads:[~2006-10-10 16:04 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-10-10 16:02 [patch 0/2] Introduce round_jiffies() to save spurious wakeups Arjan van de Ven
2006-10-10 16:03 ` Arjan van de Ven [this message]
2006-10-10 16:04   ` [patch 2/2] round_jiffies users Arjan van de Ven
2006-10-10 16:47     ` Ingo Oeser
2006-10-10 16:59       ` Arjan van de Ven
2006-10-10 22:47     ` Paul Dickson
2006-10-10 23:52       ` Arjan van de Ven
2006-10-10 18:56   ` [patch 1/2] round_jiffies infrastructure Andrew Morton
2006-10-10 20:48     ` Arjan van de Ven
2006-10-16 13:42   ` Andi Kleen
2006-10-11 17:23 ` [patch 0/2] Introduce round_jiffies() to save spurious wakeups Christoph Hellwig
2006-10-11 17:54   ` Arjan van de Ven
2006-10-12 19:02     ` Christoph Hellwig

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=1160496210.3000.310.camel@laptopd505.fenrus.org \
    --to=arjan@linux.intel.com \
    --cc=akpm@osdl.org \
    --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

all inboxes | Powered by JetHome®