linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: LKML <linux-kernel@vger.kernel.org>
Cc: Ingo Molnar <mingo@kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Paul McKenney <paulmck@linux.vnet.ibm.com>,
	Frederic Weisbecker <fweisbec@gmail.com>,
	Chris Mason <clm@fb.com>, Arjan van de Ven <arjan@infradead.org>,
	rt@linutronix.de, Rik van Riel <riel@redhat.com>,
	George Spelvin <linux@sciencehorizons.net>,
	Len Brown <lenb@kernel.org>,
	Josh Triplett <josh@joshtriplett.org>,
	Anna-Maria Gleixner <anna-maria@linutronix.de>,
	Eric Dumazet <edumazet@google.com>
Subject: [patch 4 17/22] timer: Optimize collect timers for NOHZ
Date: Mon, 04 Jul 2016 09:50:34 -0000	[thread overview]
Message-ID: <20160704094342.351296290@linutronix.de> (raw)
In-Reply-To: 20160704093956.299369787@linutronix.de

[-- Attachment #1: timer_Optimize_collect_timers_for_NOHZ.patch --]
[-- Type: text/plain, Size: 3984 bytes --]

From: Anna-Maria Gleixner <anna-maria@linutronix.de>

After a NOHZ idle sleep the wheel must be forwarded to current jiffies. There
might be expired timers so the current code loops and checks the epxired
buckets for timers. This can take quite some time for long NOHZ idle periods.

The pending bitmask in the timer base allows us to do a quick search for the
next expiring timer and therefor a fast forward of the base time which
prevents pointless long lasting loops.

For a 3 second idle sleep this reduces the catchup time from ~1ms to 5us.

Signed-off-by: Anna-Maria Gleixner <anna-maria@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Chris Mason <clm@fb.com>
Cc: Eric Dumazet <edumazet@google.com>
Cc: rt@linutronix.de
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Arjan van de Ven <arjan@infradead.org>

---
 kernel/time/timer.c |   49 +++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 41 insertions(+), 8 deletions(-)

--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -1252,8 +1252,8 @@ static void expire_timers(struct timer_b
 	}
 }
 
-static int collect_expired_timers(struct timer_base *base,
-				  struct hlist_head *heads)
+static int __collect_expired_timers(struct timer_base *base,
+				    struct hlist_head *heads)
 {
 	unsigned long clk = base->clk;
 	struct hlist_head *vec;
@@ -1279,9 +1279,9 @@ static int collect_expired_timers(struct
 
 #ifdef CONFIG_NO_HZ_COMMON
 /*
- * Find the next pending bucket of a level. Search from @offset + @clk upwards
- * and if nothing there, search from start of the level (@offset) up to
- * @offset + clk.
+ * Find the next pending bucket of a level. Search from level start (@offset)
+ * + @clk upwards and if nothing there, search from start of the level
+ * (@offset) up to @offset + clk.
  */
 static int next_pending_bucket(struct timer_base *base, unsigned offset,
 			       unsigned clk)
@@ -1298,14 +1298,14 @@ static int next_pending_bucket(struct ti
 }
 
 /*
- * Search the first expiring timer in the various clock levels.
+ * Search the first expiring timer in the various clock levels. Caller must
+ * hold base->lock.
  */
 static unsigned long __next_timer_interrupt(struct timer_base *base)
 {
 	unsigned long clk, next, adj;
 	unsigned lvl, offset = 0;
 
-	spin_lock(&base->lock);
 	next = base->clk + NEXT_TIMER_MAX_DELTA;
 	clk = base->clk;
 	for (lvl = 0; lvl < LVL_DEPTH; lvl++, offset += LVL_SIZE) {
@@ -1358,7 +1358,6 @@ static unsigned long __next_timer_interr
 		clk >>= LVL_CLK_SHIFT;
 		clk += adj;
 	}
-	spin_unlock(&base->lock);
 	return next;
 }
 
@@ -1416,7 +1415,10 @@ u64 get_next_timer_interrupt(unsigned lo
 	if (cpu_is_offline(smp_processor_id()))
 		return expires;
 
+	spin_lock(&base->lock);
 	nextevt = __next_timer_interrupt(base);
+	spin_unlock(&base->lock);
+
 	if (time_before_eq(nextevt, basej))
 		expires = basem;
 	else
@@ -1424,6 +1426,37 @@ u64 get_next_timer_interrupt(unsigned lo
 
 	return cmp_next_hrtimer_event(basem, expires);
 }
+
+static int collect_expired_timers(struct timer_base *base,
+				  struct hlist_head *heads)
+{
+	/*
+	 * NOHZ optimization. After a long idle sleep we need to forward the
+	 * base to current jiffies. Avoid a loop by searching the bitfield for
+	 * the next expiring timer.
+	 */
+	if ((long)(jiffies - base->clk) > 2) {
+		unsigned long next = __next_timer_interrupt(base);
+
+		/*
+		 * If the next timer is ahead of time forward to current
+		 * jiffies, otherwise forward to the next expiry time.
+		 */
+		if (time_after(next, jiffies)) {
+			/* The call site will increment clock! */
+			base->clk = jiffies - 1;
+			return 0;
+		}
+		base->clk = next;
+	}
+	return __collect_expired_timers(base, heads);
+}
+#else
+static inline int collect_expired_timers(struct timer_base *base,
+					 struct hlist_head *heads)
+{
+	return __collect_expired_timers(base, heads);
+}
 #endif
 
 /*

  parent reply	other threads:[~2016-07-04  9:54 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-07-04  9:50 [patch 4 00/22] timer: Refactor the timer wheel Thomas Gleixner
2016-07-04  9:50 ` [patch 4 01/22] timer: Make pinned a timer property Thomas Gleixner
2016-07-07  8:39   ` [tip:timers/core] timers: Make 'pinned' " tip-bot for Thomas Gleixner
2016-07-04  9:50 ` [patch 4 02/22] x86/apic/uv: Initialize timer as pinned Thomas Gleixner
2016-07-07  8:40   ` [tip:timers/core] timers, x86/apic/uv: Initialize the UV heartbeat " tip-bot for Thomas Gleixner
2016-07-04  9:50 ` [patch 4 03/22] x86/mce: Initialize " Thomas Gleixner
2016-07-07  8:40   ` [tip:timers/core] timers, x86/mce: Initialize MCE restart " tip-bot for Thomas Gleixner
2016-07-04  9:50 ` [patch 4 04/22] cpufreq/powernv: Initialize " Thomas Gleixner
2016-07-07  8:41   ` [tip:timers/core] timers, cpufreq/powernv: Initialize the gpstate " tip-bot for Thomas Gleixner
2016-07-04  9:50 ` [patch 4 05/22] driver/net/ethernet/tile: Initialize " Thomas Gleixner
2016-07-07  8:41   ` [tip:timers/core] timers, driver/net/ethernet/tile: Initialize the egress " tip-bot for Thomas Gleixner
2016-07-04  9:50 ` [patch 4 06/22] drivers/tty/metag_da: Initialize " Thomas Gleixner
2016-07-07  8:42   ` [tip:timers/core] timers, drivers/tty/metag_da: Initialize the poll " tip-bot for Thomas Gleixner
2016-07-04  9:50 ` [patch 4 07/22] drivers/tty/mips_ejtag: Initialize " Thomas Gleixner
2016-07-07  8:42   ` [tip:timers/core] timers, drivers/tty/mips_ejtag: Initialize the poll " tip-bot for Thomas Gleixner
2016-07-04  9:50 ` [patch 4 08/22] net/ipv4/inet: Initialize timers " Thomas Gleixner
2016-07-07  8:43   ` [tip:timers/core] timers, net/ipv4/inet: Initialize connection request " tip-bot for Thomas Gleixner
2016-07-04  9:50 ` [patch 4 09/22] timer: Remove mod_timer_pinned Thomas Gleixner
2016-07-07  8:43   ` [tip:timers/core] timers: Remove the deprecated mod_timer_pinned() API tip-bot for Thomas Gleixner
2016-07-04  9:50 ` [patch 4 10/22] signal: Use hrtimer for sigtimedwait Thomas Gleixner
2016-07-07  8:43   ` [tip:timers/core] signals: Use hrtimer for sigtimedwait() tip-bot for Thomas Gleixner
2016-07-04  9:50 ` [patch 4 11/22] hlist: Add hlist_is_singular_node() helper Thomas Gleixner
2016-07-07  8:44   ` [tip:timers/core] " tip-bot for Thomas Gleixner
2016-07-04  9:50 ` [patch 4 12/22] timer: Give a few structs and members proper names Thomas Gleixner
2016-07-07  8:44   ` [tip:timers/core] timers: " tip-bot for Thomas Gleixner
2016-07-04  9:50 ` [patch 4 13/22] timer: Reduce the CPU index space to 256k Thomas Gleixner
2016-07-07  8:45   ` [tip:timers/core] timers: " tip-bot for Thomas Gleixner
2016-07-04  9:50 ` [patch 4 14/22] timer: Switch to a non cascading wheel Thomas Gleixner
2016-07-07  8:45   ` [tip:timers/core] timers: Switch to a non-cascading wheel tip-bot for Thomas Gleixner
2016-08-11 15:21   ` [patch 4 14/22] timer: Switch to a non cascading wheel Jouni Malinen
2016-08-11 20:25     ` [PREEMPT-RT] " rcochran
2016-08-13  9:12       ` Jouni Malinen
2016-08-16  9:46         ` Richard Cochran
2016-08-16 14:35           ` Eric Dumazet
2016-08-17  9:05           ` Jouni Malinen
2016-08-17  9:23             ` rcochran
2016-08-12 17:50     ` Rik van Riel
2016-08-12 19:14       ` Paul E. McKenney
2016-08-16  8:55         ` Richard Cochran
2016-08-16  7:57       ` Richard Cochran
2016-07-04  9:50 ` [patch 4 15/22] timer: Remove slack leftovers Thomas Gleixner
2016-07-07  8:46   ` [tip:timers/core] timers: Remove set_timer_slack() leftovers tip-bot for Thomas Gleixner
2016-07-22 11:31   ` [patch 4 15/22] timer: Remove slack leftovers Jason A. Donenfeld
2016-07-22 13:04     ` Thomas Gleixner
2016-07-22 15:18       ` Jason A. Donenfeld
2016-07-22 22:54         ` Jason A. Donenfeld
2016-07-04  9:50 ` [patch 4 16/22] timer: Move __run_timers() function Thomas Gleixner
2016-07-07  8:46   ` [tip:timers/core] timers: " tip-bot for Anna-Maria Gleixner
2016-07-04  9:50 ` Thomas Gleixner [this message]
2016-07-07  8:47   ` [tip:timers/core] timers: Optimize collect_expired_timers() for NOHZ tip-bot for Anna-Maria Gleixner
2016-07-04  9:50 ` [patch 4 18/22] tick/sched: Remove pointless empty function Thomas Gleixner
2016-07-07  8:47   ` [tip:timers/core] timers/nohz: Remove pointless tick_nohz_kick_tick() function tip-bot for Thomas Gleixner
2016-07-04  9:50 ` [patch 4 19/22] timer: Forward wheel clock whenever possible Thomas Gleixner
2016-07-07  8:48   ` [tip:timers/core] timers: Forward the " tip-bot for Thomas Gleixner
2016-07-04  9:50 ` [patch 4 20/22] timer: Only wake softirq if necessary Thomas Gleixner
2016-07-07  8:48   ` [tip:timers/core] timers: " tip-bot for Thomas Gleixner
2016-07-04  9:50 ` [patch 4 21/22] timer: Split out index calculation Thomas Gleixner
2016-07-07  8:48   ` [tip:timers/core] timers: " tip-bot for Anna-Maria Gleixner
2016-07-04  9:50 ` [patch 4 22/22] timer: Optimization for same expiry time in mod_timer() Thomas Gleixner
2016-07-07  8:49   ` [tip:timers/core] timers: Implement optimization " tip-bot for Anna-Maria Gleixner

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=20160704094342.351296290@linutronix.de \
    --to=tglx@linutronix.de \
    --cc=anna-maria@linutronix.de \
    --cc=arjan@infradead.org \
    --cc=clm@fb.com \
    --cc=edumazet@google.com \
    --cc=fweisbec@gmail.com \
    --cc=josh@joshtriplett.org \
    --cc=lenb@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux@sciencehorizons.net \
    --cc=mingo@kernel.org \
    --cc=paulmck@linux.vnet.ibm.com \
    --cc=peterz@infradead.org \
    --cc=riel@redhat.com \
    --cc=rt@linutronix.de \
    /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;
as well as URLs for NNTP newsgroup(s).