PM / Domains: Add preliminary support for cpuidle, v2

On some systems there are CPU cores located in the same power
domains as I/O devices.  Then, power can only be removed from the
domain if all I/O devices in it are not in use and the CPU core
is idle.  Add preliminary support for that to the generic PM domains
framework.

First, the platform is expected to provide a cpuidle driver with one
extra state designated for use with the generic PM domains code.
This state should be initially disabled and its exit_latency value
should be set to whatever time is needed to bring up the CPU core
itself after restoring power to it, not including the domain's
power on latency.  Its .enter() callback should point to a procedure
that will remove power from the domain containing the CPU core at
the end of the CPU power transition.

The remaining characteristics of the extra cpuidle state, referred to
as the "domain" cpuidle state below, (e.g. power usage, target
residency) should be populated in accordance with the properties of
the hardware.

Next, the platform should execute genpd_attach_cpuidle() on the PM
domain containing the CPU core.  That will cause the generic PM
domains framework to treat that domain in a special way such that:

 * When all devices in the domain have been suspended and it is about
   to be turned off, the states of the devices will be saved, but
   power will not be removed from the domain.  Instead, the "domain"
   cpuidle state will be enabled so that power can be removed from
   the domain when the CPU core is idle and the state has been chosen
   as the target by the cpuidle governor.

 * When the first I/O device in the domain is resumed and
   __pm_genpd_poweron(() is called for the first time after
   power has been removed from the domain, the "domain" cpuidle
   state will be disabled to avoid subsequent surprise power removals
   via cpuidle.

The effective exit_latency value of the "domain" cpuidle state
depends on the time needed to bring up the CPU core itself after
restoring power to it as well as on the power on latency of the
domain containing the CPU core.  Thus the "domain" cpuidle state's
exit_latency has to be recomputed every time the domain's power on
latency is updated, which may happen every time power is restored
to the domain, if the measured power on latency is greater than
the latency stored in the corresponding generic_pm_domain structure.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Reviewed-by: Kevin Hilman <khilman@ti.com>

3 files changed, 120 insertions, 1 deletions

diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c
index da1d52576ec..4b5f090fccb 100644
--- a/drivers/base/power/domain.c
+++ b/drivers/base/power/domain.c
@@ -139,6 +139,19 @@ static void genpd_set_active(struct generic_pm_domain *genpd)
 		genpd->status = GPD_STATE_ACTIVE;
 }
 
+static void genpd_recalc_cpu_exit_latency(struct generic_pm_domain *genpd)
+{
+	s64 usecs64;
+
+	if (!genpd->cpu_data)
+		return;
+
+	usecs64 = genpd->power_on_latency_ns;
+	do_div(usecs64, NSEC_PER_USEC);
+	usecs64 += genpd->cpu_data->saved_exit_latency;
+	genpd->cpu_data->idle_state->exit_latency = usecs64;
+}
+
 /**
  * __pm_genpd_poweron - Restore power to a given PM domain and its masters.
  * @genpd: PM domain to power up.
@@ -176,6 +189,13 @@ int __pm_genpd_poweron(struct generic_pm_domain *genpd)
 		return 0;
 	}
 
+	if (genpd->cpu_data) {
+		cpuidle_pause_and_lock();
+		genpd->cpu_data->idle_state->disabled = true;
+		cpuidle_resume_and_unlock();
+		goto out;
+	}
+
 	/*
 	 * The list is guaranteed not to change while the loop below is being
 	 * executed, unless one of the masters' .power_on() callbacks fiddles
@@ -215,6 +235,7 @@ int __pm_genpd_poweron(struct generic_pm_domain *genpd)
 		if (elapsed_ns > genpd->power_on_latency_ns) {
 			genpd->power_on_latency_ns = elapsed_ns;
 			genpd->max_off_time_changed = true;
+			genpd_recalc_cpu_exit_latency(genpd);
 			if (genpd->name)
 				pr_warning("%s: Power-on latency exceeded, "
 					"new value %lld ns\n", genpd->name,
@@ -222,6 +243,7 @@ int __pm_genpd_poweron(struct generic_pm_domain *genpd)
 		}
 	}
 
+ out:
 	genpd_set_active(genpd);
 
 	return 0;
@@ -455,6 +477,21 @@ static int pm_genpd_poweroff(struct generic_pm_domain *genpd)
 		}
 	}
 
+	if (genpd->cpu_data) {
+		/*
+		 * If cpu_data is set, cpuidle should turn the domain off when
+		 * the CPU in it is idle.  In that case we don't decrement the
+		 * subdomain counts of the master domains, so that power is not
+		 * removed from the current domain prematurely as a result of
+		 * cutting off the masters' power.
+		 */
+		genpd->status = GPD_STATE_POWER_OFF;
+		cpuidle_pause_and_lock();
+		genpd->cpu_data->idle_state->disabled = false;
+		cpuidle_resume_and_unlock();
+		goto out;
+	}
+
 	if (genpd->power_off) {
 		ktime_t time_start;
 		s64 elapsed_ns;
@@ -1600,6 +1637,86 @@ int __pm_genpd_remove_callbacks(struct device *dev, bool clear_td)
 }
 EXPORT_SYMBOL_GPL(__pm_genpd_remove_callbacks);
 
+int genpd_attach_cpuidle(struct generic_pm_domain *genpd, int state)
+{
+	struct cpuidle_driver *cpuidle_drv;
+	struct gpd_cpu_data *cpu_data;
+	struct cpuidle_state *idle_state;
+	int ret = 0;
+
+	if (IS_ERR_OR_NULL(genpd) || state < 0)
+		return -EINVAL;
+
+	genpd_acquire_lock(genpd);
+
+	if (genpd->cpu_data) {
+		ret = -EEXIST;
+		goto out;
+	}
+	cpu_data = kzalloc(sizeof(*cpu_data), GFP_KERNEL);
+	if (!cpu_data) {
+		ret = -ENOMEM;
+		goto out;
+	}
+	cpuidle_drv = cpuidle_driver_ref();
+	if (!cpuidle_drv) {
+		ret = -ENODEV;
+		goto out;
+	}
+	if (cpuidle_drv->state_count <= state) {
+		ret = -EINVAL;
+		goto err;
+	}
+	idle_state = &cpuidle_drv->states[state];
+	if (!idle_state->disabled) {
+		ret = -EAGAIN;
+		goto err;
+	}
+	cpu_data->idle_state = idle_state;
+	cpu_data->saved_exit_latency = idle_state->exit_latency;
+	genpd->cpu_data = cpu_data;
+	genpd_recalc_cpu_exit_latency(genpd);
+
+ out:
+	genpd_release_lock(genpd);
+	return ret;
+
+ err:
+	cpuidle_driver_unref();
+	goto out;
+}
+
+int genpd_detach_cpuidle(struct generic_pm_domain *genpd)
+{
+	struct gpd_cpu_data *cpu_data;
+	struct cpuidle_state *idle_state;
+	int ret = 0;
+
+	if (IS_ERR_OR_NULL(genpd))
+		return -EINVAL;
+
+	genpd_acquire_lock(genpd);
+
+	cpu_data = genpd->cpu_data;
+	if (!cpu_data) {
+		ret = -ENODEV;
+		goto out;
+	}
+	idle_state = cpu_data->idle_state;
+	if (!idle_state->disabled) {
+		ret = -EAGAIN;
+		goto out;
+	}
+	idle_state->exit_latency = cpu_data->saved_exit_latency;
+	cpuidle_driver_unref();
+	genpd->cpu_data = NULL;
+	kfree(cpu_data);
+
+ out:
+	genpd_release_lock(genpd);
+	return ret;
+}
+
 /* Default device callbacks for generic PM domains. */
 
 /**
diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c
index 04e4b7674a4..0132706251d 100644
--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -265,6 +265,7 @@ static void poll_idle_init(struct cpuidle_driver *drv)
 	state->power_usage = -1;
 	state->flags = 0;
 	state->enter = poll_idle;
+	state->disabled = false;
 }
 #else
 static void poll_idle_init(struct cpuidle_driver *drv) {}
diff --git a/drivers/cpuidle/governors/menu.c b/drivers/cpuidle/governors/menu.c
index 8391d93f57d..5b1f2c372c1 100644
--- a/drivers/cpuidle/governors/menu.c
+++ b/drivers/cpuidle/governors/menu.c
@@ -281,6 +281,7 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev)
 	 * unless the timer is happening really really soon.
 	 */
 	if (data->expected_us > 5 &&
+	    !drv->states[CPUIDLE_DRIVER_STATE_START].disabled &&
 		dev->states_usage[CPUIDLE_DRIVER_STATE_START].disable == 0)
 		data->last_state_idx = CPUIDLE_DRIVER_STATE_START;
 
@@ -292,7 +293,7 @@ static int menu_select(struct cpuidle_driver *drv, struct cpuidle_device *dev)
 		struct cpuidle_state *s = &drv->states[i];
 		struct cpuidle_state_usage *su = &dev->states_usage[i];
 
-		if (su->disable)
+		if (s->disabled || su->disable)
 			continue;
 		if (s->target_residency > data->predicted_us)
 			continue;