From: Peter Zijlstra <a.p.zijlstra@chello.nl>
To: linux-kernel@vger.kernel.org
Cc: mingo@elte.hu, efault@gmx.de, vatsa@in.ibm.com,
Peter Zijlstra <a.p.zijlstra@chello.nl>
Subject: [PATCH 8/8] use avg_vruntime for task placement
Date: Fri, 24 Oct 2008 11:06:19 +0200 [thread overview]
Message-ID: <20081024091109.998783999@chello.nl> (raw)
In-Reply-To: 20081024090611.936552213@chello.nl
[-- Attachment #1: sched-placement.patch --]
[-- Type: text/plain, Size: 1919 bytes --]
Allow to use the avg_vruntime for task placement.
The pro: its the 'fair' place to insert tasks.
The con: its an extra division.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
kernel/sched.c | 9 +++++++--
kernel/sched_fair.c | 7 ++++++-
kernel/sched_features.h | 1 +
3 files changed, 14 insertions(+), 3 deletions(-)
Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -1847,8 +1847,13 @@ void set_task_cpu(struct task_struct *p,
schedstat_inc(p, se.nr_forced2_migrations);
}
#endif
- p->se.vruntime -= old_cfsrq->min_vruntime -
- new_cfsrq->min_vruntime;
+ if (sched_feat(AVG_VRUNTIME)) {
+ p->se.vruntime -=
+ avg_vruntime(old_cfsrq) - avg_vruntime(new_cfsrq);
+ } else {
+ p->se.vruntime -=
+ old_cfsrq->min_vruntime - new_cfsrq->min_vruntime;
+ }
__set_task_cpu(p, new_cpu);
}
Index: linux-2.6/kernel/sched_fair.c
===================================================================
--- linux-2.6.orig/kernel/sched_fair.c
+++ linux-2.6/kernel/sched_fair.c
@@ -719,7 +719,12 @@ static void check_spread(struct cfs_rq *
static void
place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int initial)
{
- u64 vruntime = cfs_rq->min_vruntime;
+ u64 vruntime;
+
+ if (sched_feat(AVG_VRUNTIME))
+ vruntime = avg_vruntime(cfs_rq);
+ else
+ vruntime = cfs_rq->min_vruntime;
/*
* The 'current' period is already promised to the current tasks,
Index: linux-2.6/kernel/sched_features.h
===================================================================
--- linux-2.6.orig/kernel/sched_features.h
+++ linux-2.6/kernel/sched_features.h
@@ -11,4 +11,5 @@ SCHED_FEAT(ASYM_GRAN, 1)
SCHED_FEAT(LB_BIAS, 1)
SCHED_FEAT(LB_WAKEUP_UPDATE, 1)
SCHED_FEAT(ASYM_EFF_LOAD, 1)
+SCHED_FEAT(AVG_VRUNTIME, 0)
SCHED_FEAT(WAKEUP_OVERLAP, 0)
--
next prev parent reply other threads:[~2008-10-24 10:01 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-10-24 9:06 [PATCH 0/8] scheduler patches Peter Zijlstra
2008-10-24 9:06 ` [PATCH 1/8] sched: fix a find_busiest_group buglet Peter Zijlstra
2008-10-24 9:06 ` [PATCH 2/8] sched: more accurate min_vruntime accounting Peter Zijlstra
2008-10-24 9:06 ` [PATCH 3/8] sched: weaken sync hint Peter Zijlstra
2008-10-24 9:06 ` [PATCH 4/8] sched: re-instate vruntime based wakeup preemption Peter Zijlstra
2008-10-24 9:06 ` [PATCH 5/8] sched: virtual time buddy preemption Peter Zijlstra
2008-10-24 9:06 ` [PATCH 6/8] sched: avg_vruntime Peter Zijlstra
2008-10-29 15:48 ` Peter Zijlstra
2008-11-01 18:13 ` Fabio Checconi
2008-10-24 9:06 ` [PATCH 7/8] sched: non-zero lag renice Peter Zijlstra
2008-10-24 17:47 ` Chris Friesen
2008-10-24 20:28 ` Peter Zijlstra
2008-10-24 21:13 ` Chris Friesen
2008-10-24 9:06 ` Peter Zijlstra [this message]
2008-10-24 10:26 ` [PATCH 0/8] scheduler patches Ingo Molnar
2008-10-24 10:29 ` 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=20081024091109.998783999@chello.nl \
--to=a.p.zijlstra@chello.nl \
--cc=efault@gmx.de \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=vatsa@in.ibm.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