From: Viresh Kumar <viresh.kumar@linaro.org>
To: pjt@google.com, paul.mckenney@linaro.org, tglx@linutronix.de,
tj@kernel.org, suresh.b.siddha@intel.com, venki@google.com,
mingo@redhat.com, peterz@infradead.org, rostedt@goodmis.org
Cc: linaro-kernel@lists.linaro.org, robin.randhawa@arm.com,
Steve.Bannister@arm.com, Liviu.Dudau@arm.com,
charles.garcia-tobin@arm.com, Arvind.Chauhan@arm.com,
linux-rt-users@vger.kernel.org, linux-kernel@vger.kernel.org,
Viresh Kumar <viresh.kumar@linaro.org>,
Jens Axboe <axboe@kernel.dk>
Subject: [PATCH V3 6/7] block: queue work on any cpu
Date: Mon, 18 Mar 2013 20:53:28 +0530 [thread overview]
Message-ID: <9aa157216e60de7d245128d42fb896918184ae8c.1363617402.git.viresh.kumar@linaro.org> (raw)
In-Reply-To: <cover.1363617402.git.viresh.kumar@linaro.org>
In-Reply-To: <cover.1363617402.git.viresh.kumar@linaro.org>
block layer uses workqueues for multiple purposes. There is no real dependency
of scheduling these on the cpu which scheduled them.
On a idle system, it is observed that and idle cpu wakes up many times just to
service this work. It would be better if we can schedule it on a cpu which isn't
idle to save on power.
By idle cpu (from scheduler's perspective) we mean:
- Current task is idle task
- nr_running == 0
- wake_list is empty
This patch replaces schedule_work() and queue_[delayed_]work() with
queue_[delayed_]work_on_any_cpu() siblings.
These routines would look for the closest (via scheduling domains) non-idle cpu
(non-idle from schedulers perspective). If the current cpu is not idle or all
cpus are idle, work will be scheduled on local cpu.
Cc: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
---
block/blk-core.c | 6 +++---
block/blk-ioc.c | 2 +-
block/genhd.c | 9 ++++++---
3 files changed, 10 insertions(+), 7 deletions(-)
diff --git a/block/blk-core.c b/block/blk-core.c
index 186603b..14ae74f 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -225,7 +225,7 @@ static void blk_delay_work(struct work_struct *work)
void blk_delay_queue(struct request_queue *q, unsigned long msecs)
{
if (likely(!blk_queue_dead(q)))
- queue_delayed_work(kblockd_workqueue, &q->delay_work,
+ queue_delayed_work_on_any_cpu(kblockd_workqueue, &q->delay_work,
msecs_to_jiffies(msecs));
}
EXPORT_SYMBOL(blk_delay_queue);
@@ -2852,14 +2852,14 @@ EXPORT_SYMBOL_GPL(blk_rq_prep_clone);
int kblockd_schedule_work(struct request_queue *q, struct work_struct *work)
{
- return queue_work(kblockd_workqueue, work);
+ return queue_work_on_any_cpu(kblockd_workqueue, work);
}
EXPORT_SYMBOL(kblockd_schedule_work);
int kblockd_schedule_delayed_work(struct request_queue *q,
struct delayed_work *dwork, unsigned long delay)
{
- return queue_delayed_work(kblockd_workqueue, dwork, delay);
+ return queue_delayed_work_on_any_cpu(kblockd_workqueue, dwork, delay);
}
EXPORT_SYMBOL(kblockd_schedule_delayed_work);
diff --git a/block/blk-ioc.c b/block/blk-ioc.c
index 9c4bb82..2eefeb1 100644
--- a/block/blk-ioc.c
+++ b/block/blk-ioc.c
@@ -144,7 +144,7 @@ void put_io_context(struct io_context *ioc)
if (atomic_long_dec_and_test(&ioc->refcount)) {
spin_lock_irqsave(&ioc->lock, flags);
if (!hlist_empty(&ioc->icq_list))
- schedule_work(&ioc->release_work);
+ queue_work_on_any_cpu(system_wq, &ioc->release_work);
else
free_ioc = true;
spin_unlock_irqrestore(&ioc->lock, flags);
diff --git a/block/genhd.c b/block/genhd.c
index a1ed52a..4bdb735 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -1488,9 +1488,11 @@ static void __disk_unblock_events(struct gendisk *disk, bool check_now)
intv = disk_events_poll_jiffies(disk);
set_timer_slack(&ev->dwork.timer, intv / 4);
if (check_now)
- queue_delayed_work(system_freezable_wq, &ev->dwork, 0);
+ queue_delayed_work_on_any_cpu(system_freezable_wq, &ev->dwork,
+ 0);
else if (intv)
- queue_delayed_work(system_freezable_wq, &ev->dwork, intv);
+ queue_delayed_work_on_any_cpu(system_freezable_wq, &ev->dwork,
+ intv);
out_unlock:
spin_unlock_irqrestore(&ev->lock, flags);
}
@@ -1626,7 +1628,8 @@ static void disk_check_events(struct disk_events *ev,
intv = disk_events_poll_jiffies(disk);
if (!ev->block && intv)
- queue_delayed_work(system_freezable_wq, &ev->dwork, intv);
+ queue_delayed_work_on_any_cpu(system_freezable_wq, &ev->dwork,
+ intv);
spin_unlock_irq(&ev->lock);
--
1.7.12.rc2.18.g61b472e
next prev parent reply other threads:[~2013-03-18 15:23 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-18 15:23 [PATCH V3 0/7] Create sched_select_cpu() and use it for workqueues Viresh Kumar
2013-03-18 15:23 ` [PATCH V3 1/7] sched: Create sched_select_cpu() to give preferred CPU for power saving Viresh Kumar
2013-03-18 15:39 ` Frederic Weisbecker
2013-03-18 15:44 ` Viresh Kumar
2013-03-18 15:57 ` Frederic Weisbecker
2013-03-19 5:15 ` Viresh Kumar
2013-03-19 12:30 ` Peter Zijlstra
2013-03-19 12:52 ` Viresh Kumar
2013-03-19 13:22 ` Viresh Kumar
2013-03-18 15:23 ` [PATCH V3 2/7] timer: hrtimer: Don't check idle_cpu() before calling get_nohz_timer_target() Viresh Kumar
2013-03-18 15:23 ` [PATCH V3 3/7] workqueue: Add helpers to schedule work on any cpu Viresh Kumar
2013-03-19 5:15 ` Viresh Kumar
2013-03-19 13:23 ` Viresh Kumar
2013-03-21 0:12 ` Tejun Heo
2013-03-21 10:57 ` Viresh Kumar
2013-03-21 18:29 ` Tejun Heo
2013-03-28 18:13 ` Tejun Heo
2013-03-29 2:39 ` Viresh Kumar
2013-03-29 7:27 ` Viresh Kumar
2013-03-29 17:40 ` Tejun Heo
2013-03-29 17:56 ` Tejun Heo
2013-03-30 3:30 ` Viresh Kumar
2013-03-18 15:23 ` [PATCH V3 4/7] PHYLIB: queue " Viresh Kumar
2013-03-18 17:33 ` David Miller
2013-03-18 15:23 ` [PATCH V3 5/7] mmc: " Viresh Kumar
2013-03-19 7:58 ` Ulf Hansson
2013-03-22 17:09 ` Chris Ball
2013-03-22 17:26 ` Chris Ball
2013-03-22 17:27 ` Viresh Kumar
2013-03-22 17:30 ` Chris Ball
2013-03-18 15:23 ` Viresh Kumar [this message]
2013-03-22 15:05 ` [PATCH V3 6/7] block: " Jens Axboe
2013-03-23 6:44 ` Viresh Kumar
2013-03-18 15:23 ` [PATCH V3 7/7] fbcon: " Viresh Kumar
2013-03-19 5:00 ` [PATCH V3 0/7] Create sched_select_cpu() and use it for workqueues Viresh Kumar
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=9aa157216e60de7d245128d42fb896918184ae8c.1363617402.git.viresh.kumar@linaro.org \
--to=viresh.kumar@linaro.org \
--cc=Arvind.Chauhan@arm.com \
--cc=Liviu.Dudau@arm.com \
--cc=Steve.Bannister@arm.com \
--cc=axboe@kernel.dk \
--cc=charles.garcia-tobin@arm.com \
--cc=linaro-kernel@lists.linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-rt-users@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=paul.mckenney@linaro.org \
--cc=peterz@infradead.org \
--cc=pjt@google.com \
--cc=robin.randhawa@arm.com \
--cc=rostedt@goodmis.org \
--cc=suresh.b.siddha@intel.com \
--cc=tglx@linutronix.de \
--cc=tj@kernel.org \
--cc=venki@google.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;
as well as URLs for NNTP newsgroup(s).