From: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
To: LKML <linux-kernel@vger.kernel.org>,
Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
Andrew Morton <akpm@linux-foundation.org>,
Ingo Molnar <mingo@elte.hu>, Steven Rostedt <rostedt@goodmis.org>,
Thomas Gleixner <tglx@linutronix.de>,
Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
Tony Lindgren <tony@atomide.com>, Mike Galbraith <efault@gmx.de>
Subject: [RFC PATCH 11/11] sched: fair sleepers for timer and interactive
Date: Thu, 26 Aug 2010 14:09:19 -0400 [thread overview]
Message-ID: <20100826181342.149176927@efficios.com> (raw)
In-Reply-To: 20100826180908.648103531@efficios.com
[-- Attachment #1: sched-fair-sleeper-for-timer-and-interactive.patch --]
[-- Type: text/plain, Size: 1812 bytes --]
Add FAIR_SLEEPERS_TIMER and FAIR_SLEEPERS_INTERACTIVE tuning knobs.
Turning off FAIR_SLEEPERS and turning on FAIR_SLEEPERS_INTERACTIVE does a pretty
good job for interactivity.
Setting FAIR_SLEEPERS to off helps keeping the min_vruntime value somewhere
between MIN_vruntime and max_vruntime rather than somewhere at the right of all
running threads.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
---
kernel/sched_fair.c | 5 ++++-
kernel/sched_features.h | 4 +++-
2 files changed, 7 insertions(+), 2 deletions(-)
Index: linux-2.6-lttng.git/kernel/sched_fair.c
===================================================================
--- linux-2.6-lttng.git.orig/kernel/sched_fair.c
+++ linux-2.6-lttng.git/kernel/sched_fair.c
@@ -748,7 +748,10 @@ place_entity(struct cfs_rq *cfs_rq, stru
vruntime += sched_vslice(cfs_rq, se);
/* sleeps up to a single latency don't count. */
- if (sched_feat(FAIR_SLEEPERS) && !initial) {
+ if (!initial
+ && (sched_feat(FAIR_SLEEPERS)
+ || (sched_feat(FAIR_SLEEPERS_TIMER) && se->timer)
+ || (sched_feat(FAIR_SLEEPERS_INTERACTIVE) && se->interactive))) {
unsigned long thresh = sysctl_sched_latency;
/*
Index: linux-2.6-lttng.git/kernel/sched_features.h
===================================================================
--- linux-2.6-lttng.git.orig/kernel/sched_features.h
+++ linux-2.6-lttng.git/kernel/sched_features.h
@@ -3,7 +3,9 @@
* them to run sooner, but does not allow tons of sleepers to
* rip the spread apart.
*/
-SCHED_FEAT(FAIR_SLEEPERS, 1)
+SCHED_FEAT(FAIR_SLEEPERS, 1) /* Apply to all wakeups */
+SCHED_FEAT(FAIR_SLEEPERS_INTERACTIVE, 0) /* Selects interactive wakeups */
+SCHED_FEAT(FAIR_SLEEPERS_TIMER, 0) /* Selects timer-driven wakeups */
SCHED_FEAT(GENTLE_FAIR_SLEEPERS, 1)
/*
next prev parent reply other threads:[~2010-08-26 18:14 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-08-26 18:09 [RFC PATCH 00/11] sched: CFS low-latency features Mathieu Desnoyers
2010-08-26 18:09 ` [RFC PATCH 01/11] sched: fix string comparison in features Mathieu Desnoyers
2010-08-26 18:09 ` [RFC PATCH 02/11] sched: debug spread check account for nr_running Mathieu Desnoyers
2010-08-26 18:09 ` [RFC PATCH 03/11] sched: FAIR_SLEEPERS feature Mathieu Desnoyers
2010-08-26 18:09 ` [RFC PATCH 04/11] sched: debug cleanup place entity Mathieu Desnoyers
2010-08-26 18:09 ` [RFC PATCH 05/11] sched buddy enable buddy logic starting at 2 running threads Mathieu Desnoyers
2010-08-26 18:09 ` [RFC PATCH 06/11] sched: dynamic min_vruntime Mathieu Desnoyers
2010-08-26 18:09 ` [RFC PATCH 07/11] sched rename struct task in_iowait field to sched_in_iowait Mathieu Desnoyers
2010-08-26 18:09 ` [RFC PATCH 08/11] sched input interactivity-driven next buddy Mathieu Desnoyers
2010-08-26 18:09 ` [RFC PATCH 09/11] sched: timer-driven " Mathieu Desnoyers
2010-08-27 18:02 ` [RFC PATCH 09/11] sched: timer-driven next buddy (update) Mathieu Desnoyers
2010-08-27 18:14 ` Thomas Gleixner
2010-08-26 18:09 ` [RFC PATCH 10/11] sched: fork expedited Mathieu Desnoyers
2010-08-26 18:09 ` Mathieu Desnoyers [this message]
2010-08-26 18:57 ` [RFC PATCH 00/11] sched: CFS low-latency features Peter Zijlstra
2010-08-26 21:25 ` Thomas Gleixner
2010-08-26 22:22 ` Thomas Gleixner
2010-08-26 23:09 ` Mathieu Desnoyers
2010-08-26 23:36 ` Mathieu Desnoyers
2010-08-27 7:38 ` Peter Zijlstra
2010-08-27 15:23 ` Mathieu Desnoyers
2010-08-27 8:43 ` Thomas Gleixner
2010-08-27 15:50 ` Mathieu Desnoyers
2010-08-27 7:37 ` Peter Zijlstra
2010-08-27 15:21 ` Mathieu Desnoyers
2010-08-27 15:41 ` Peter Zijlstra
2010-08-27 16:09 ` Mathieu Desnoyers
2010-08-27 17:27 ` Peter Zijlstra
2010-08-27 18:32 ` Mathieu Desnoyers
2010-08-27 19:23 ` Peter Zijlstra
2010-08-27 19:57 ` Mathieu Desnoyers
2010-08-31 15:02 ` Mathieu Desnoyers
2010-08-26 23:18 ` Paul E. McKenney
2010-08-26 23:28 ` Mathieu Desnoyers
2010-08-26 23:38 ` Paul E. McKenney
2010-08-26 23:53 ` Mathieu Desnoyers
2010-08-27 0:09 ` Paul E. McKenney
2010-08-27 15:18 ` Mathieu Desnoyers
2010-08-27 15:20 ` Thomas Gleixner
2010-08-27 15:30 ` Mathieu Desnoyers
2010-08-27 15:41 ` Peter Zijlstra
2010-08-26 23:49 ` Mathieu Desnoyers
2010-08-27 7:42 ` Peter Zijlstra
2010-08-27 8:19 ` Mike Galbraith
2010-08-27 15:43 ` Mathieu Desnoyers
2010-08-27 18:38 ` Mathieu Desnoyers
2010-08-28 7:33 ` Mike Galbraith
2010-08-27 10:47 ` Indan Zupancic
2010-08-27 10:58 ` 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=20100826181342.149176927@efficios.com \
--to=mathieu.desnoyers@efficios.com \
--cc=akpm@linux-foundation.org \
--cc=efault@gmx.de \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
--cc=tony@atomide.com \
--cc=torvalds@linux-foundation.org \
/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