From: Paul Turner <pjt@google.com>
To: linux-kernel@vger.kernel.org
Cc: Venki Pallipadi <venki@google.com>,
Srivatsa Vaddagiri <vatsa@in.ibm.com>,
Vincent Guittot <vincent.guittot@linaro.org>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Nikunj A Dadhania <nikunj@linux.vnet.ibm.com>,
Mike Galbraith <efault@gmx.de>,
Kamalesh Babulal <kamalesh@linux.vnet.ibm.com>,
Ben Segall <bsegall@google.com>, Ingo Molnar <mingo@elte.hu>,
"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
Morten Rasmussen <Morten.Rasmussen@arm.com>,
Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>
Subject: [PATCH 16/16] sched: introduce temporary FAIR_GROUP_SCHED dependency for load-tracking
Date: Wed, 27 Jun 2012 19:24:15 -0700 [thread overview]
Message-ID: <20120628022415.30496.18884.stgit@kitami.mtv.corp.google.com> (raw)
In-Reply-To: <20120628022413.30496.32798.stgit@kitami.mtv.corp.google.com>
While per-entity load-tracking is generally useful, beyond computing shares
distribution, e.g.
runnable based load-balance (in progress), governors, power-management, etc
These facilities are not yet consumers of this data. This may be trivially
reverted when the information is required; but avoid paying the overhead for
calculations we will not use until then.
Signed-off-by: Paul Turner <pjt@google.com>
---
include/linux/sched.h | 8 +++++++-
kernel/sched/fair.c | 14 +++++++++++---
kernel/sched/sched.h | 9 ++++++++-
3 files changed, 26 insertions(+), 5 deletions(-)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index a65c097..7c0c72d 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1205,7 +1205,13 @@ struct sched_entity {
/* rq "owned" by this entity/group: */
struct cfs_rq *my_q;
#endif
-#ifdef CONFIG_SMP
+/*
+ * Load-tracking only depends on SMP, FAIR_GROUP_SCHED dependency below may be
+ * removed when useful for applications beyond shares distribution (e.g.
+ * load-balance).
+ */
+#if defined(CONFIG_SMP) && defined(CONFIG_FAIR_GROUP_SCHED)
+ /* Per-entity load-tracking */
struct sched_avg avg;
#endif
};
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 3d1aaa8..0b6e2df 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -882,7 +882,8 @@ static inline void update_cfs_shares(struct cfs_rq *cfs_rq)
}
#endif /* CONFIG_FAIR_GROUP_SCHED */
-#ifdef CONFIG_SMP
+/* Only depends on SMP, FAIR_GROUP_SCHED may be removed when useful in lb */
+#if defined(CONFIG_SMP) && defined(CONFIG_FAIR_GROUP_SCHED)
/*
* We choose a half-life close to 1 scheduling period.
* Note: The tables below are dependent on this value.
@@ -3214,6 +3215,12 @@ unlock:
}
/*
+ * Load-tracking only depends on SMP, FAIR_GROUP_SCHED dependency below may be
+ * removed when useful for applications beyond shares distribution (e.g.
+ * load-balance).
+ */
+#ifdef CONFIG_FAIR_GROUP_SCHED
+/*
* Called immediately before a task is migrated to a new cpu; task_cpu(p) and
* cfs_rq_of(p) references at time of call are still valid and identify the
* previous cpu. However, the caller only guarantees p->pi_lock is held; no
@@ -3235,7 +3242,7 @@ migrate_task_rq_fair(struct task_struct *p, int next_cpu) {
atomic64_add(se->avg.load_avg_contrib, &cfs_rq->removed_load);
}
}
-
+#endif
#endif /* CONFIG_SMP */
@@ -5927,8 +5934,9 @@ const struct sched_class fair_sched_class = {
#ifdef CONFIG_SMP
.select_task_rq = select_task_rq_fair,
+#ifdef CONFIG_FAIR_GROUP_SCHED
.migrate_task_rq = migrate_task_rq_fair,
-
+#endif
.rq_online = rq_online_fair,
.rq_offline = rq_offline_fair,
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index bb76895..d920943 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -225,6 +225,12 @@ struct cfs_rq {
#endif
#ifdef CONFIG_SMP
+/*
+ * Load-tracking only depends on SMP, FAIR_GROUP_SCHED dependency below may be
+ * removed when useful for applications beyond shares distribution (e.g.
+ * load-balance).
+ */
+#ifdef CONFIG_FAIR_GROUP_SCHED
/*
* CFS Load tracking
* Under CFS, load is tracked on a per-entity basis and aggregated up.
@@ -234,7 +240,8 @@ struct cfs_rq {
u64 runnable_load_avg, blocked_load_avg;
atomic64_t decay_counter, removed_load;
u64 last_decay;
-
+#endif /* CONFIG_FAIR_GROUP_SCHED */
+/* These always depend on CONFIG_FAIR_GROUP_SCHED */
#ifdef CONFIG_FAIR_GROUP_SCHED
u32 tg_runnable_contrib, tg_usage_contrib;
u64 tg_load_contrib;
next prev parent reply other threads:[~2012-06-28 2:57 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-28 2:24 [PATCH 00/16] Series short description Paul Turner
2012-06-28 2:24 ` [PATCH 09/16] sched: normalize tg load contributions against runnable time Paul Turner
2012-06-29 7:26 ` Namhyung Kim
2012-07-04 19:48 ` Peter Zijlstra
2012-07-06 11:52 ` Peter Zijlstra
2012-07-12 1:08 ` Andre Noll
2012-07-12 0:02 ` Paul Turner
2012-07-06 12:23 ` Peter Zijlstra
2012-06-28 2:24 ` [PATCH 06/16] sched: account for blocked load waking back up Paul Turner
2012-06-28 2:24 ` [PATCH 02/16] sched: maintain per-rq runnable averages Paul Turner
2012-06-28 2:24 ` [PATCH 01/16] sched: track the runnable average on a per-task entitiy basis Paul Turner
2012-06-28 6:06 ` Namhyung Kim
2012-07-12 0:14 ` Paul Turner
2012-07-04 15:32 ` Peter Zijlstra
2012-07-12 0:12 ` Paul Turner
2012-06-28 2:24 ` [PATCH 04/16] sched: maintain the load contribution of blocked entities Paul Turner
2012-06-29 1:27 ` Namhyung Kim
2012-06-28 2:24 ` [PATCH 07/16] sched: aggregate total task_group load Paul Turner
2012-06-28 2:24 ` [PATCH 05/16] sched: add an rq migration call-back to sched_class Paul Turner
2012-06-29 1:32 ` Namhyung Kim
2012-06-28 2:24 ` [PATCH 08/16] sched: compute load contribution by a group entity Paul Turner
2012-06-28 2:24 ` [PATCH 03/16] sched: aggregate load contributed by task entities on parenting cfs_rq Paul Turner
2012-06-28 6:33 ` Namhyung Kim
2012-07-04 15:28 ` Peter Zijlstra
2012-07-06 14:53 ` Peter Zijlstra
2012-07-09 9:15 ` Ingo Molnar
2012-06-28 2:24 ` [PATCH 11/16] sched: replace update_shares weight distribution with per-entity computation Paul Turner
2012-06-28 2:24 ` Paul Turner [this message]
2012-06-28 2:24 ` [PATCH 12/16] sched: refactor update_shares_cpu() -> update_blocked_avgs() Paul Turner
2012-06-29 7:28 ` Namhyung Kim
2012-07-12 0:03 ` Paul Turner
2012-07-05 11:58 ` Peter Zijlstra
2012-07-12 0:11 ` Paul Turner
2012-07-12 14:40 ` Peter Zijlstra
2012-06-28 2:24 ` [PATCH 15/16] sched: implement usage tracking Paul Turner
2012-06-28 2:24 ` [PATCH 13/16] sched: update_cfs_shares at period edge Paul Turner
2012-06-28 2:24 ` [PATCH 10/16] sched: maintain runnable averages across throttled periods Paul Turner
2012-06-28 2:24 ` [PATCH 14/16] sched: make __update_entity_runnable_avg() fast Paul Turner
2012-07-04 15:41 ` Peter Zijlstra
2012-07-04 17:20 ` Peter Zijlstra
2012-07-09 20:18 ` Benjamin Segall
2012-07-10 10:51 ` Peter Zijlstra
2012-07-12 0:15 ` Paul Turner
2012-07-12 14:30 ` Peter Zijlstra
2012-07-04 16:51 ` Peter Zijlstra
2012-08-23 14:14 [patch 00/16] sched: per-entity load-tracking pjt
2012-08-23 14:14 ` [patch 16/16] sched: introduce temporary FAIR_GROUP_SCHED dependency for load-tracking pjt
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=20120628022415.30496.18884.stgit@kitami.mtv.corp.google.com \
--to=pjt@google.com \
--cc=Morten.Rasmussen@arm.com \
--cc=a.p.zijlstra@chello.nl \
--cc=bsegall@google.com \
--cc=efault@gmx.de \
--cc=kamalesh@linux.vnet.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=nikunj@linux.vnet.ibm.com \
--cc=paulmck@linux.vnet.ibm.com \
--cc=svaidy@linux.vnet.ibm.com \
--cc=vatsa@in.ibm.com \
--cc=venki@google.com \
--cc=vincent.guittot@linaro.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;
as well as URLs for NNTP newsgroup(s).