LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Paul Turner <pjt@google.com>
To: linux-kernel@vger.kernel.org
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Ingo Molnar <mingo@elte.hu>, Mike Galbraith <efault@gmx.de>
Subject: [patch 5/5] sched: avoid expensive initial update_cfs_load()
Date: Fri, 21 Jan 2011 20:45:03 -0800	[thread overview]
Message-ID: <20110122044852.102126037@google.com> (raw)
In-Reply-To: <20110122044458.058531078@google.com>

[-- Attachment #1: sched-fix_first_update_load.patch --]
[-- Type: text/plain, Size: 1648 bytes --]

Since cfs->{load_stamp,load_last} are zero-initalized the initial load update
will consider the delta to be 'since the beginning of time'.

This results in a lot of pointless divisions to bring this large period to be
within the sysctl_sched_shares_window.

Fix this by initializing load_stamp to be 1 at cfs_rq initialization, this
allows for an initial load_stamp > load_last which then lets standard idle
truncation proceed.

We avoid spinning (and slightly improve consistency) by fixing delta to be 
[period - 1] in this path resulting in a slightly more predictable shares ramp.
(Previously the amount of idle time preserved by the overflow would range between
[period/2,period-1].)

Signed-off-by: Paul Turner <pjt@google.com>

---
 kernel/sched.c      |    2 ++
 kernel/sched_fair.c |    1 +
 2 files changed, 3 insertions(+)

Index: tip3/kernel/sched.c
===================================================================
--- tip3.orig/kernel/sched.c
+++ tip3/kernel/sched.c
@@ -7796,6 +7796,8 @@ static void init_cfs_rq(struct cfs_rq *c
 	INIT_LIST_HEAD(&cfs_rq->tasks);
 #ifdef CONFIG_FAIR_GROUP_SCHED
 	cfs_rq->rq = rq;
+	/* allow initial update_cfs_load() to truncate */
+	cfs_rq->load_stamp = 1;
 #endif
 	cfs_rq->min_vruntime = (u64)(-(1LL << 20));
 }
Index: tip3/kernel/sched_fair.c
===================================================================
--- tip3.orig/kernel/sched_fair.c
+++ tip3/kernel/sched_fair.c
@@ -732,6 +732,7 @@ static void update_cfs_load(struct cfs_r
 	    now - cfs_rq->load_last > 4 * period) {
 		cfs_rq->load_period = 0;
 		cfs_rq->load_avg = 0;
+		delta = period - 1;
 	}
 
 	cfs_rq->load_stamp = now;



  parent reply	other threads:[~2011-01-22  5:05 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-01-22  4:44 [patch 0/5] scheduler fixlets Paul Turner
2011-01-22  4:44 ` [patch 1/5] sched: fix sign under-flows in wake_affine Paul Turner
2011-01-26 12:09   ` [tip:sched/core] sched: Fix " tip-bot for Paul Turner
2011-01-22  4:45 ` [patch 2/5] sched: (cleanup) remove redundant cfs_rq checks Paul Turner
2011-01-26 12:10   ` [tip:sched/core] sched: Fix/remove " tip-bot for Paul Turner
2011-01-22  4:45 ` [patch 3/5] sched: simplify update_cfs_shares parameters Paul Turner
2011-01-26 12:11   ` [tip:sched/core] sched: Simplify " tip-bot for Paul Turner
2011-01-22  4:45 ` [patch 4/5] sched: use rq->clock_task instead of rq->clock for maintaining load averages Paul Turner
2011-01-26 12:10   ` [tip:sched/core] sched: Use rq->clock_task instead of rq->clock for correctly " tip-bot for Paul Turner
2011-01-22  4:45 ` Paul Turner [this message]
2011-01-26 12:11   ` [tip:sched/core] sched: Avoid expensive initial update_cfs_load() tip-bot for Paul Turner
2011-01-26 12:36     ` Peter Zijlstra
2011-01-26 12:45   ` [tip:sched/core] sched: Avoid expensive initial update_cfs_load(), on UP too tip-bot for Peter Zijlstra
2011-01-27 11:58   ` tip-bot for Peter Zijlstra
2011-01-24 10:17 ` [patch 0/5] scheduler fixlets Peter Zijlstra

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=20110122044852.102126037@google.com \
    --to=pjt@google.com \
    --cc=a.p.zijlstra@chello.nl \
    --cc=efault@gmx.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --subject='Re: [patch 5/5] sched: avoid expensive initial update_cfs_load()' \
    /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

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).