Merge branch 'for-3.11' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq

Pull workqueue changes from Tejun Heo:
"Surprisingly, Lai and I didn't break too many things implementing
custom pools and stuff last time around and there aren't any follow-up
changes necessary at this point.

The only change in this pull request is Viresh's patches to make some
per-cpu workqueues to behave as unbound workqueues dependent on a boot
param whose default can be configured via a config option. This leads
to higher processing overhead / lower bandwidth as more work items are
bounced across CPUs; however, it can lead to noticeable powersave in
certain configurations - ~10% w/ idlish constant workload on a
big.LITTLE configuration according to Viresh.

This is because per-cpu workqueues interfere with how the scheduler
perceives whether or not each CPU is idle by forcing pinned tasks on
them, which makes the scheduler's power-aware scheduling decisions
less effective.

Its effectiveness is likely less pronounced on homogenous
configurations and this type of optimization can probably be made
automatic; however, the changes are pretty minimal and the affected
workqueues are clearly marked, so it's an easy gain for some
configurations for the time being with pretty unintrusive changes."

* 'for-3.11' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:
fbcon: queue work on power efficient wq
block: queue work on power efficient wq
PHYLIB: queue work on system_power_efficient_wq
workqueue: Add system wide power_efficient workqueues
workqueues: Introduce new flag WQ_POWER_EFFICIENT for power oriented workqueues

Documentation/kernel-parameters.txt

···
 			that this also can be controlled per-workqueue for
 			workqueues visible under /sys/bus/workqueue/.
 
+	workqueue.power_efficient
+			Per-cpu workqueues are generally preferred because
+			they show better performance thanks to cache
+			locality; unfortunately, per-cpu workqueues tend to
+			be more power hungry than unbound workqueues.
+
+			Enabling this makes the per-cpu workqueues which
+			were observed to contribute significantly to power
+			consumption unbound, leading to measurably lower
+			power usage at the cost of small performance
+			overhead.
+
+			The default value of this parameter is determined by
+			the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
+
 	x2apic_phys	[X86-64,APIC] Use x2apic physical mode instead of
 			default x2apic cluster mode on platforms
 			supporting x2apic.

block/blk-core.c

···
 
 	/* used for unplugging and affects IO latency/throughput - HIGHPRI */
 	kblockd_workqueue = alloc_workqueue("kblockd",
-					    WQ_MEM_RECLAIM | WQ_HIGHPRI, 0);
+					    WQ_MEM_RECLAIM | WQ_HIGHPRI |
+					    WQ_POWER_EFFICIENT, 0);
 	if (!kblockd_workqueue)
 		panic("Failed to create kblockd\n");
 

block/blk-ioc.c

···
 	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(system_power_efficient_wq,
+					&ioc->release_work);
 		else
 			free_ioc = true;
 		spin_unlock_irqrestore(&ioc->lock, flags);

block/genhd.c

···
 	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(system_freezable_power_efficient_wq,
+				&ev->dwork, 0);
 	else if (intv)
-		queue_delayed_work(system_freezable_wq, &ev->dwork, intv);
+		queue_delayed_work(system_freezable_power_efficient_wq,
+				&ev->dwork, intv);
 out_unlock:
 	spin_unlock_irqrestore(&ev->lock, flags);
 }
···
 	spin_lock_irq(&ev->lock);
 	ev->clearing |= mask;
 	if (!ev->block)
-		mod_delayed_work(system_freezable_wq, &ev->dwork, 0);
+		mod_delayed_work(system_freezable_power_efficient_wq,
+				&ev->dwork, 0);
 	spin_unlock_irq(&ev->lock);
 }
 
···
 
 	intv = disk_events_poll_jiffies(disk);
 	if (!ev->block && intv)
-		queue_delayed_work(system_freezable_wq, &ev->dwork, intv);
+		queue_delayed_work(system_freezable_power_efficient_wq,
+				&ev->dwork, intv);
 
 	spin_unlock_irq(&ev->lock);
 

drivers/net/phy/phy.c

···
 {
 	phydev->adjust_state = handler;
 
-	schedule_delayed_work(&phydev->state_queue, HZ);
+	queue_delayed_work(system_power_efficient_wq, &phydev->state_queue, HZ);
 }
 
 /**
···
 	disable_irq_nosync(irq);
 	atomic_inc(&phydev->irq_disable);
 
-	schedule_work(&phydev->phy_queue);
+	queue_work(system_power_efficient_wq, &phydev->phy_queue);
 
 	return IRQ_HANDLED;
 }
···
 
 	/* reschedule state queue work to run as soon as possible */
 	cancel_delayed_work_sync(&phydev->state_queue);
-	schedule_delayed_work(&phydev->state_queue, 0);
+	queue_delayed_work(system_power_efficient_wq, &phydev->state_queue, 0);
 
 	return;
 
···
 	if (err < 0)
 		phy_error(phydev);
 
-	schedule_delayed_work(&phydev->state_queue, PHY_STATE_TIME * HZ);
+	queue_delayed_work(system_power_efficient_wq, &phydev->state_queue,
+			PHY_STATE_TIME * HZ);
 }
 
 static inline void mmd_phy_indirect(struct mii_bus *bus, int prtad, int devad,

drivers/video/console/fbcon.c

···
 	struct fb_info *info = (struct fb_info *) dev_addr;
 	struct fbcon_ops *ops = info->fbcon_par;
 
-	schedule_work(&info->queue);
+	queue_work(system_power_efficient_wq, &info->queue);
 	mod_timer(&ops->cursor_timer, jiffies + HZ/5);
 }
 

include/linux/workqueue.h

···
 	WQ_CPU_INTENSIVE	= 1 << 5, /* cpu instensive workqueue */
 	WQ_SYSFS		= 1 << 6, /* visible in sysfs, see wq_sysfs_register() */
 
+	/*
+	 * Per-cpu workqueues are generally preferred because they tend to
+	 * show better performance thanks to cache locality.  Per-cpu
+	 * workqueues exclude the scheduler from choosing the CPU to
+	 * execute the worker threads, which has an unfortunate side effect
+	 * of increasing power consumption.
+	 *
+	 * The scheduler considers a CPU idle if it doesn't have any task
+	 * to execute and tries to keep idle cores idle to conserve power;
+	 * however, for example, a per-cpu work item scheduled from an
+	 * interrupt handler on an idle CPU will force the scheduler to
+	 * excute the work item on that CPU breaking the idleness, which in
+	 * turn may lead to more scheduling choices which are sub-optimal
+	 * in terms of power consumption.
+	 *
+	 * Workqueues marked with WQ_POWER_EFFICIENT are per-cpu by default
+	 * but become unbound if workqueue.power_efficient kernel param is
+	 * specified.  Per-cpu workqueues which are identified to
+	 * contribute significantly to power-consumption are identified and
+	 * marked with this flag and enabling the power_efficient mode
+	 * leads to noticeable power saving at the cost of small
+	 * performance disadvantage.
+	 *
+	 * http://thread.gmane.org/gmane.linux.kernel/1480396
+	 */
+	WQ_POWER_EFFICIENT	= 1 << 7,
+
 	__WQ_DRAINING		= 1 << 16, /* internal: workqueue is draining */
 	__WQ_ORDERED		= 1 << 17, /* internal: workqueue is ordered */
 
···
  *
  * system_freezable_wq is equivalent to system_wq except that it's
  * freezable.
+ *
+ * *_power_efficient_wq are inclined towards saving power and converted
+ * into WQ_UNBOUND variants if 'wq_power_efficient' is enabled; otherwise,
+ * they are same as their non-power-efficient counterparts - e.g.
+ * system_power_efficient_wq is identical to system_wq if
+ * 'wq_power_efficient' is disabled.  See WQ_POWER_EFFICIENT for more info.
  */
 extern struct workqueue_struct *system_wq;
 extern struct workqueue_struct *system_long_wq;
 extern struct workqueue_struct *system_unbound_wq;
 extern struct workqueue_struct *system_freezable_wq;
+extern struct workqueue_struct *system_power_efficient_wq;
+extern struct workqueue_struct *system_freezable_power_efficient_wq;
 
 static inline struct workqueue_struct * __deprecated __system_nrt_wq(void)
 {

kernel/power/Kconfig

···
 	bool
 	depends on PM
 
+config WQ_POWER_EFFICIENT_DEFAULT
+	bool "Enable workqueue power-efficient mode by default"
+	depends on PM
+	default n
+	help
+	  Per-cpu workqueues are generally preferred because they show
+	  better performance thanks to cache locality; unfortunately,
+	  per-cpu workqueues tend to be more power hungry than unbound
+	  workqueues.
+
+	  Enabling workqueue.power_efficient kernel parameter makes the
+	  per-cpu workqueues which were observed to contribute
+	  significantly to power consumption unbound, leading to measurably
+	  lower power usage at the cost of small performance overhead.
+
+	  This config option determines whether workqueue.power_efficient
+	  is enabled by default.
+
+	  If in doubt, say N.
+
 config PM_GENERIC_DOMAINS_SLEEP
 	def_bool y
 	depends on PM_SLEEP && PM_GENERIC_DOMAINS

kernel/workqueue.c

···
 static bool wq_disable_numa;
 module_param_named(disable_numa, wq_disable_numa, bool, 0444);
 
+/* see the comment above the definition of WQ_POWER_EFFICIENT */
+#ifdef CONFIG_WQ_POWER_EFFICIENT_DEFAULT
+static bool wq_power_efficient = true;
+#else
+static bool wq_power_efficient;
+#endif
+
+module_param_named(power_efficient, wq_power_efficient, bool, 0444);
+
 static bool wq_numa_enabled;		/* unbound NUMA affinity enabled */
 
 /* buf for wq_update_unbound_numa_attrs(), protected by CPU hotplug exclusion */
···
 EXPORT_SYMBOL_GPL(system_unbound_wq);
 struct workqueue_struct *system_freezable_wq __read_mostly;
 EXPORT_SYMBOL_GPL(system_freezable_wq);
+struct workqueue_struct *system_power_efficient_wq __read_mostly;
+EXPORT_SYMBOL_GPL(system_power_efficient_wq);
+struct workqueue_struct *system_freezable_power_efficient_wq __read_mostly;
+EXPORT_SYMBOL_GPL(system_freezable_power_efficient_wq);
 
 static int worker_thread(void *__worker);
 static void copy_workqueue_attrs(struct workqueue_attrs *to,
···
 	struct workqueue_struct *wq;
 	struct pool_workqueue *pwq;
 
+	/* see the comment above the definition of WQ_POWER_EFFICIENT */
+	if ((flags & WQ_POWER_EFFICIENT) && wq_power_efficient)
+		flags |= WQ_UNBOUND;
+
 	/* allocate wq and format name */
 	if (flags & WQ_UNBOUND)
 		tbl_size = wq_numa_tbl_len * sizeof(wq->numa_pwq_tbl[0]);
···
 					    WQ_UNBOUND_MAX_ACTIVE);
 	system_freezable_wq = alloc_workqueue("events_freezable",
 					      WQ_FREEZABLE, 0);
+	system_power_efficient_wq = alloc_workqueue("events_power_efficient",
+					      WQ_POWER_EFFICIENT, 0);
+	system_freezable_power_efficient_wq = alloc_workqueue("events_freezable_power_efficient",
+					      WQ_FREEZABLE | WQ_POWER_EFFICIENT,
+					      0);
 	BUG_ON(!system_wq || !system_highpri_wq || !system_long_wq ||
-	       !system_unbound_wq || !system_freezable_wq);
+	       !system_unbound_wq || !system_freezable_wq ||
+	       !system_power_efficient_wq ||
+	       !system_freezable_power_efficient_wq);
 	return 0;
 }
 early_initcall(init_workqueues);