LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
To: LKML <linux-kernel@vger.kernel.org>, Ingo Molnar <mingo@elte.hu>
Cc: Dmitry Adamushko <dmitry.adamushko@gmail.com>,
	Mike Galbraith <efault@gmx.de>,
	Dhaval Giani <dhaval@linux.vnet.ibm.com>,
	Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>
Subject: [RFC/PATCH 16/17] sched: fair: bound tardiness
Date: Sun, 09 Mar 2008 18:09:06 +0100	[thread overview]
Message-ID: <20080309170929.637205000@chello.nl> (raw)
In-Reply-To: <20080309170850.256853000@chello.nl>

[-- Attachment #1: sched-fair-slice-limit.patch --]
[-- Type: text/plain, Size: 1366 bytes --]

EEVDF guarantees a O(1) tardiness exceeding the deadline, this tardines comes
from the slice length. So by limiting the slice length when there is
competition, we limit the amount by which the deadline can be exceeded.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched_fair.c |   14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

Index: linux-2.6-2/kernel/sched_fair.c
===================================================================
--- linux-2.6-2.orig/kernel/sched_fair.c
+++ linux-2.6-2/kernel/sched_fair.c
@@ -584,15 +584,21 @@ static u64 sched_slice(struct cfs_rq *cf
 {
 	u64 slice = __sched_period(cfs_rq->nr_running);
 
-	/*
-	 * FIXME: curious 'hack' to make it boot - when the tick is started we
-	 * hit this with the init_task, which is not actually enqueued.
-	 */
 	if (unlikely(rq_of(cfs_rq)->load.weight <= se->load.weight))
 		goto out;
 
 	slice = calc_delta_weight(slice, se);
 
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	/*
+	 * EEVDF guarantees a O(1) tardiness exceeding the deadline, this
+	 * tardines comes from the slice length. So by limiting the slice
+	 * length when there is competition, we limit the amount by which the
+	 * deadline can be exceeded.
+	 */
+	if (rq_of(cfs_rq)->cfs_root.nr_queued)
+		slice = min_t(u64, slice, sysctl_sched_min_granularity);
+#endif
 out:
 	return slice;
 }

--


  parent reply	other threads:[~2008-03-09 17:17 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-03-09 17:08 [RFC/PATCH 00/17] Current sched patch stack Peter Zijlstra
2008-03-09 17:08 ` [RFC/PATCH 01/17] sched: mix tasks and groups Peter Zijlstra
2008-04-01 12:12   ` Srivatsa Vaddagiri
2008-04-01 12:05     ` Peter Zijlstra
2008-03-09 17:08 ` [RFC/PATCH 02/17] sched: allow the CFS group scheduler to have multiple levels Peter Zijlstra
2008-03-12 14:47   ` Dhaval Giani
2008-03-09 17:08 ` [RFC/PATCH 03/17] sched: rt-group: full hierarchy support for the rt groups Peter Zijlstra
2008-03-12  8:25   ` Dhaval Giani
     [not found]     ` <661de9470803120129l66497656o948c13c6c0c26766@mail.gmail.com>
2008-03-12  9:29       ` Peter Zijlstra
2008-03-12 13:34         ` Balbir Singh
2008-03-12 13:39           ` Dhaval Giani
2008-03-12 14:08         ` Balbir Singh
2008-03-09 17:08 ` [RFC/PATCH 04/17] sched: fair-group: SMP-nice for group scheduling Peter Zijlstra
2008-03-09 17:08 ` [RFC/PATCH 05/17] sched: rt-group: optimize dequeue_rt_stack Peter Zijlstra
2008-03-09 17:08 ` [RFC/PATCH 06/17] sched: fair-group scheduling vs latency Peter Zijlstra
2008-03-09 17:08 ` [RFC/PATCH 07/17] sched: fair-group: de-couple load-balancing from the rb-trees Peter Zijlstra
2008-03-09 17:08 ` [RFC/PATCH 08/17] sched: fair-group: single RQ approach Peter Zijlstra
2008-03-09 17:08 ` [RFC/PATCH 09/17] sched: fair: remove moved_group() Peter Zijlstra
2008-03-09 17:09 ` [RFC/PATCH 10/17] sched: remove sysctl_sched_batch_wakeup_granularity Peter Zijlstra
2008-03-09 17:09 ` [RFC/PATCH 11/17] sched: fair: optimize sched_slice() Peter Zijlstra
2008-03-09 17:09 ` [RFC/PATCH 12/17] sched: fair: cfs_root_rq Peter Zijlstra
2008-03-09 17:09 ` [RFC/PATCH 13/17] sched: fair: calc_delta_weight() Peter Zijlstra
2008-03-09 17:09 ` [RFC/PATCH 14/17] sched: fair: avg_vruntime Peter Zijlstra
2008-03-09 17:09 ` [RFC/PATCH 15/17] sched: fair-group: EEVDF scheduling Peter Zijlstra
2008-03-09 17:09 ` Peter Zijlstra [this message]
2008-03-09 17:09 ` [RFC/PATCH 17/17] sched: fair: optimize the elegebility test 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=20080309170929.637205000@chello.nl \
    --to=a.p.zijlstra@chello.nl \
    --cc=dhaval@linux.vnet.ibm.com \
    --cc=dmitry.adamushko@gmail.com \
    --cc=efault@gmx.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=vatsa@linux.vnet.ibm.com \
    --subject='Re: [RFC/PATCH 16/17] sched: fair: bound tardiness' \
    /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).