All of 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>,
	Bharata B Rao <bharata@linux.vnet.ibm.com>,
	Dhaval Giani <dhaval.giani@gmail.com>,
	Balbir Singh <balbir@linux.vnet.ibm.com>,
	Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>,
	Srivatsa Vaddagiri <vatsa@in.ibm.com>,
	Kamalesh Babulal <kamalesh@linux.vnet.ibm.com>,
	Ingo Molnar <mingo@elte.hu>, Pavel Emelyanov <xemul@openvz.org>
Subject: [patch 08/15] sched: migrate throttled tasks on HOTPLUG
Date: Tue, 22 Mar 2011 20:03:34 -0700	[thread overview]
Message-ID: <20110323030449.434436052@google.com> (raw)
In-Reply-To: 20110323030326.789836913@google.com

[-- Attachment #1: sched-bwc-migrate_dead.patch --]
[-- Type: text/plain, Size: 1688 bytes --]

Throttled tasks are invisisble to cpu-offline since they are not eligible for
selection by pick_next_task().  The regular 'escape' path for a thread that is
blocked at offline is via ttwu->select_task_rq, however this will not handle a
throttled group since there are no individual thread wakeups on an unthrottle.

Resolve this by unthrottling offline cpus so that threads can be migrated.

Signed-off-by: Paul Turner <pjt@google.com>
---
 kernel/sched.c |   27 +++++++++++++++++++++++++++
 1 file changed, 27 insertions(+)

Index: tip/kernel/sched.c
===================================================================
--- tip.orig/kernel/sched.c
+++ tip/kernel/sched.c
@@ -6119,6 +6119,30 @@ static void calc_global_load_remove(stru
 	rq->calc_load_active = 0;
 }
 
+#ifdef CONFIG_CFS_BANDWIDTH
+static void unthrottle_offline_cfs_rqs(struct rq *rq)
+{
+	struct cfs_rq *cfs_rq;
+
+	for_each_leaf_cfs_rq(rq, cfs_rq) {
+		if (!cfs_rq->quota_enabled)
+			continue;
+
+		/*
+		 * clock_task is not advancing so we just need to make sure
+		 * there's some valid quota amount
+		 */
+		cfs_rq->quota_remaining = tg_cfs_bandwidth(cfs_rq->tg)->quota;
+		if (cfs_rq_throttled(cfs_rq))
+			unthrottle_cfs_rq(cfs_rq);
+	}
+}
+#else
+static void unthrottle_offline_cfs_rqs(struct rq *rq)
+{
+}
+#endif
+
 /*
  * Migrate all tasks from the rq, sleeping tasks will be migrated by
  * try_to_wake_up()->select_task_rq().
@@ -6144,6 +6168,9 @@ static void migrate_tasks(unsigned int d
 	 */
 	rq->stop = NULL;
 
+	/* Ensure any throttled groups are reachable by pick_next_task */
+	unthrottle_offline_cfs_rqs(rq);
+
 	for ( ; ; ) {
 		/*
 		 * There's this thread running, bail when that's the only



  parent reply	other threads:[~2011-03-23  3:10 UTC|newest]

Thread overview: 63+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-03-23  3:03 [patch 00/15] CFS Bandwidth Control V5 Paul Turner
2011-03-23  3:03 ` [patch 01/15] sched: introduce primitives to account for CFS bandwidth tracking Paul Turner
2011-03-24 12:38   ` Kamalesh Babulal
2011-04-05 13:28   ` Peter Zijlstra
2011-03-23  3:03 ` [patch 02/15] sched: validate CFS quota hierarchies Paul Turner
2011-03-23 10:39   ` torbenh
2011-03-23 20:49     ` Paul Turner
2011-03-24  6:31   ` Bharata B Rao
2011-04-08 17:01     ` Peter Zijlstra
2011-03-29  6:57   ` Hidetoshi Seto
2011-04-04 23:10     ` Paul Turner
2011-04-05 13:28   ` Peter Zijlstra
2011-03-23  3:03 ` [patch 03/15] sched: accumulate per-cfs_rq cpu usage Paul Turner
2011-04-05 13:28   ` Peter Zijlstra
2011-04-06 20:44     ` Paul Turner
2011-04-05 13:28   ` Peter Zijlstra
2011-04-06 20:47     ` Paul Turner
2011-03-23  3:03 ` [patch 04/15] sched: throttle cfs_rq entities which exceed their local quota Paul Turner
2011-03-23  5:09   ` Mike Galbraith
2011-03-23 20:53     ` Paul Turner
2011-03-24  6:36   ` Bharata B Rao
2011-03-24  7:40     ` Paul Turner
2011-04-05 13:28   ` Peter Zijlstra
2011-04-05 23:15     ` Paul Turner
2011-03-23  3:03 ` [patch 05/15] sched: unthrottle cfs_rq(s) who ran out of quota at period refresh Paul Turner
2011-04-05 13:28   ` Peter Zijlstra
2011-04-05 13:33     ` Peter Zijlstra
2011-04-05 13:28   ` Peter Zijlstra
2011-04-05 13:28   ` Peter Zijlstra
2011-03-23  3:03 ` [patch 06/15] sched: allow for positional tg_tree walks Paul Turner
2011-03-23  3:03 ` [patch 07/15] sched: prevent interactions between throttled entities and load-balance Paul Turner
2011-04-05 13:28   ` Peter Zijlstra
2011-03-23  3:03 ` Paul Turner [this message]
2011-04-05 13:28   ` [patch 08/15] sched: migrate throttled tasks on HOTPLUG Peter Zijlstra
2011-04-06  2:31     ` Paul Turner
2011-03-23  3:03 ` [patch 09/15] sched: add exports tracking cfs bandwidth control statistics Paul Turner
2011-04-05 13:28   ` Peter Zijlstra
2011-03-23  3:03 ` [patch 10/15] sched: (fixlet) dont update shares twice on on_rq parent Paul Turner
2011-04-05 13:28   ` Peter Zijlstra
2011-03-23  3:03 ` [patch 11/15] sched: hierarchical task accounting for SCHED_OTHER Paul Turner
2011-04-05 13:28   ` Peter Zijlstra
2011-03-23  3:03 ` [patch 12/15] sched: maintain throttled rqs as a list Paul Turner
2011-04-22  2:50   ` Hidetoshi Seto
2011-04-24 21:23     ` Paul Turner
2011-03-23  3:03 ` [patch 13/15] sched: expire slack quota using generation counters Paul Turner
2011-04-05 13:28   ` Peter Zijlstra
2011-04-06  7:22     ` Paul Turner
2011-04-06  8:15       ` Peter Zijlstra
2011-04-06 11:26       ` Peter Zijlstra
2011-03-23  3:03 ` [patch 14/15] sched: return unused quota on voluntary sleep Paul Turner
2011-04-05 13:28   ` Peter Zijlstra
2011-04-06  2:25     ` Paul Turner
2011-03-23  3:03 ` [patch 15/15] sched: add documentation for bandwidth control Paul Turner
2011-03-24  6:38   ` Bharata B Rao
2011-03-24 16:12 ` [patch 00/15] CFS Bandwidth Control V5 Bharata B Rao
2011-03-31  7:57 ` Xiao Guangrong
2011-04-04 23:10   ` Paul Turner
2011-04-05 13:28 ` Peter Zijlstra
2011-05-20  2:12 ` Test for CFS Bandwidth Control V6 Xiao Guangrong
2011-05-24  0:53   ` Hidetoshi Seto
2011-05-24  7:56     ` Xiao Guangrong
2011-06-08  2:54     ` Paul Turner
2011-06-08  5:55       ` Hidetoshi Seto

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=20110323030449.434436052@google.com \
    --to=pjt@google.com \
    --cc=a.p.zijlstra@chello.nl \
    --cc=balbir@linux.vnet.ibm.com \
    --cc=bharata@linux.vnet.ibm.com \
    --cc=dhaval.giani@gmail.com \
    --cc=kamalesh@linux.vnet.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=svaidy@linux.vnet.ibm.com \
    --cc=vatsa@in.ibm.com \
    --cc=xemul@openvz.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.