[PATCH 0/3] thermal: rework core to improve userspace interaction

[PATCH 0/3] thermal: rework core to improve userspace interaction

[Eduardo Valentin]

Hello Rui, linux-pm

Please consider these three patches in the thermal core to improve
the interaction with userspace.

The first is already in its second version. It avoids reconfiguring
monitor period. Now the thermal core configures the monitor only
after handling all trip points.

The second is about improving emul_temp. The background here is to
allow using emul_temp, even if the thermal zone is not fully setup,
with a missing .get_temp().

The third is to improve hot trip points handling. Hot trip points
are described as notification entry points. However, we do very
little on them. This patch adds a uevent to propagate the event
to userspace. Today, we rely on thermal zone driver. I believe
having the same message coming on every thermal zone makes more
sense. However, I did not remove the .notify() callback, and it
should behave the same.

BR,

Eduardo Valentin (3):
  thermal: setup monitor only once after handling trips
  thermal: rework core to allow emul_temp to be treated as regular temp
  thermal: improve hot trip handling

 drivers/thermal/thermal_core.c | 96 ++++++++++++++++++++++++++++++++++--------
 1 file changed, 78 insertions(+), 18 deletions(-)

-- 
2.5.0

[PATCHv2 1/3] thermal: setup monitor only once after handling trips

[Eduardo Valentin]

Instead of changing the monitoring setup every time after
handling each trip, this patch simplifies the monitoring
setup by moving the setup call to a place where all
trips have been treated already.

Cc: Zhang Rui <rui.zhang@intel.com>
Cc: linux-pm@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
---
V1->V2:
	fixed typo
---
 drivers/thermal/thermal_core.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/thermal/thermal_core.c b/drivers/thermal/thermal_core.c
index d9e525c..8fa82c0 100644
--- a/drivers/thermal/thermal_core.c
+++ b/drivers/thermal/thermal_core.c
@@ -457,11 +457,6 @@ static void handle_thermal_trip(struct thermal_zone_device *tz, int trip)
 		handle_critical_trips(tz, trip, type);
 	else
 		handle_non_critical_trips(tz, trip, type);
-	/*
-	 * Alright, we handled this trip successfully.
-	 * So, start monitoring again.
-	 */
-	monitor_thermal_zone(tz);
 }
 
 /**
@@ -547,6 +542,12 @@ void thermal_zone_device_update(struct thermal_zone_device *tz)
 
 	for (count = 0; count < tz->trips; count++)
 		handle_thermal_trip(tz, count);
+
+	/*
+	 * Alright, we handled these trips successfully.
+	 * So, start monitoring again.
+	 */
+	monitor_thermal_zone(tz);
 }
 EXPORT_SYMBOL_GPL(thermal_zone_device_update);
 
-- 
2.5.0

Re: [PATCHv2 1/3] thermal: setup monitor only once after handling trips

[Pandruvada, Srinivas]

On Tue, 2015-12-15 at 19:49 -0800, Eduardo Valentin wrote:
> Instead of changing the monitoring setup every time after
> handling each trip, this patch simplifies the monitoring
> setup by moving the setup call to a place where all
> trips have been treated already.
> 
Looks good to me.

Thanks,
Srinivas

> Cc: Zhang Rui <rui.zhang@intel.com>
> Cc: linux-pm@vger.kernel.org
> Cc: linux-kernel@vger.kernel.org
> Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
> ---
> V1->V2:
> 	fixed typo
> ---
>  drivers/thermal/thermal_core.c | 11 ++++++-----
>  1 file changed, 6 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/thermal/thermal_core.c b/drivers/thermal/thermal_core.c
> index d9e525c..8fa82c0 100644
> --- a/drivers/thermal/thermal_core.c
> +++ b/drivers/thermal/thermal_core.c
> @@ -457,11 +457,6 @@ static void handle_thermal_trip(struct thermal_zone_device *tz, int trip)
>  		handle_critical_trips(tz, trip, type);
>  	else
>  		handle_non_critical_trips(tz, trip, type);
> -	/*
> -	 * Alright, we handled this trip successfully.
> -	 * So, start monitoring again.
> -	 */
> -	monitor_thermal_zone(tz);
>  }
>  
>  /**
> @@ -547,6 +542,12 @@ void thermal_zone_device_update(struct thermal_zone_device *tz)
>  
>  	for (count = 0; count < tz->trips; count++)
>  		handle_thermal_trip(tz, count);
> +
> +	/*
> +	 * Alright, we handled these trips successfully.
> +	 * So, start monitoring again.
> +	 */
> +	monitor_thermal_zone(tz);
>  }
>  EXPORT_SYMBOL_GPL(thermal_zone_device_update);
>  

[PATCH 2/3] thermal: rework core to allow emul_temp to be treated as regular temp

[Eduardo Valentin]

Change the way we handle emul_temp. This is to allow
emul_temp to be used as input to exercise thermal core
properly. Now, even if .get_temp is not available,
the emul_temp can be used to emulate temperature.

Cc: Zhang Rui <rui.zhang@intel.com>
Cc: linux-pm@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
---
 drivers/thermal/thermal_core.c | 33 ++++++++++++++++++++-------------
 1 file changed, 20 insertions(+), 13 deletions(-)

diff --git a/drivers/thermal/thermal_core.c b/drivers/thermal/thermal_core.c
index 8fa82c0..f826589 100644
--- a/drivers/thermal/thermal_core.c
+++ b/drivers/thermal/thermal_core.c
@@ -472,18 +472,25 @@ static void handle_thermal_trip(struct thermal_zone_device *tz, int trip)
 int thermal_zone_get_temp(struct thermal_zone_device *tz, int *temp)
 {
 	int ret = -EINVAL;
-	int count;
-	int crit_temp = INT_MAX;
-	enum thermal_trip_type type;
+	int crit_temp = INT_MAX, real_temp = crit_temp - 1;
 
-	if (!tz || IS_ERR(tz) || !tz->ops->get_temp)
-		goto exit;
+	if (!tz || IS_ERR(tz))
+		return ret;
 
 	mutex_lock(&tz->lock);
 
-	ret = tz->ops->get_temp(tz, temp);
+	/* Allow emulation if .get_temp is still not available */
+	if (tz->ops->get_temp) {
+		ret = tz->ops->get_temp(tz, temp);
+		if (!ret)
+			real_temp = *temp;
+	}
 
 	if (IS_ENABLED(CONFIG_THERMAL_EMULATION) && tz->emul_temperature) {
+		enum thermal_trip_type type;
+		int count;
+
+		ret = 0;
 		for (count = 0; count < tz->trips; count++) {
 			ret = tz->ops->get_trip_type(tz, count, &type);
 			if (!ret && type == THERMAL_TRIP_CRITICAL) {
@@ -498,17 +505,17 @@ int thermal_zone_get_temp(struct thermal_zone_device *tz, int *temp)
 		 * is below the critical temperature so that the emulation code
 		 * cannot hide critical conditions.
 		 */
-		if (!ret && *temp < crit_temp)
+		if (!ret && real_temp < crit_temp)
 			*temp = tz->emul_temperature;
 	}
  
 	mutex_unlock(&tz->lock);
-exit:
+
 	return ret;
 }
 EXPORT_SYMBOL_GPL(thermal_zone_get_temp);
 
-static void update_temperature(struct thermal_zone_device *tz)
+static int update_temperature(struct thermal_zone_device *tz)
 {
 	int temp, ret;
 
@@ -518,7 +525,7 @@ static void update_temperature(struct thermal_zone_device *tz)
 			dev_warn(&tz->device,
 				 "failed to read out thermal zone (%d)\n",
 				 ret);
-		return;
+		return ret;
 	}
 
 	mutex_lock(&tz->lock);
@@ -529,17 +536,17 @@ static void update_temperature(struct thermal_zone_device *tz)
 	trace_thermal_temperature(tz);
 	dev_dbg(&tz->device, "last_temperature=%d, current_temperature=%d\n",
 				tz->last_temperature, tz->temperature);
+
+	return 0;
 }
 
 void thermal_zone_device_update(struct thermal_zone_device *tz)
 {
 	int count;
 
-	if (!tz->ops->get_temp)
+	if (update_temperature(tz))
 		return;
 
-	update_temperature(tz);
-
 	for (count = 0; count < tz->trips; count++)
 		handle_thermal_trip(tz, count);
 
-- 
2.5.0

[PATCH 3/3] thermal: improve hot trip handling

[Eduardo Valentin]

The idea is to add the choice to be notified only when temperature
crosses trip points. The trip points affected are the non-passive
trip points.

It will check last temperature and current temperature against
the trip point temperature and its hysteresis.
In case the check shows temperature has changed enought indicating
a trip point crossing, a uevent will be sent to userspace.

The uevent contains the thermal zone type, the current temperature,
the last temperature and the trip point in which the current temperature
now resides.

The behavior of ops->notify() callback remains the same.

Cc: Zhang Rui <rui.zhang@intel.com>
Cc: linux-pm@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
---
 drivers/thermal/thermal_core.c | 52 ++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 52 insertions(+)

diff --git a/drivers/thermal/thermal_core.c b/drivers/thermal/thermal_core.c
index f826589..924cccf 100644
--- a/drivers/thermal/thermal_core.c
+++ b/drivers/thermal/thermal_core.c
@@ -423,6 +423,56 @@ static void handle_non_critical_trips(struct thermal_zone_device *tz,
 		       def_governor->throttle(tz, trip);
 }
 
+static void thermal_tripped_notify(struct thermal_zone_device *tz,
+				   int trip, enum thermal_trip_type trip_type,
+				   int trip_temp)
+{
+	char tuv_name[THERMAL_NAME_LENGTH + 15], tuv_temp[25],
+		tuv_ltemp[25], tuv_trip[25], tuv_type[25];
+	char *msg[6] = { tuv_name, tuv_temp, tuv_ltemp, tuv_trip, tuv_type,
+			NULL };
+	int upper_trip_hyst, upper_trip_temp, trip_hyst = 0;
+	int ret = 0;
+
+	snprintf(tuv_name, sizeof(tuv_name), "THERMAL_ZONE=%s", tz->type);
+	snprintf(tuv_temp, sizeof(tuv_temp), "TEMP=%d", tz->temperature);
+	snprintf(tuv_ltemp, sizeof(tuv_ltemp), "LAST_TEMP=%d",
+		 tz->last_temperature);
+	snprintf(tuv_trip, sizeof(tuv_trip), "TRIP=%d", trip);
+	snprintf(tuv_type, sizeof(tuv_type), "TRIP_TYPE=%d", trip_type);
+
+	mutex_lock(&tz->lock);
+
+	/* crossing up */
+	if (tz->last_temperature < trip_temp && trip_temp < tz->temperature)
+		kobject_uevent_env(&tz->device.kobj, KOBJ_CHANGE, msg);
+
+	if (tz->ops->get_trip_hyst)
+		tz->ops->get_trip_hyst(tz, trip, &trip_hyst);
+
+	/* crossing down, check for hyst */
+	trip_temp -= trip_hyst;
+	if (tz->last_temperature > trip_temp && trip_temp > tz->temperature) {
+		snprintf(tuv_trip, sizeof(tuv_trip), "TRIP=%d", trip - 1);
+		kobject_uevent_env(&tz->device.kobj, KOBJ_CHANGE, msg);
+	}
+
+	ret = tz->ops->get_trip_temp(tz, trip + 1, &upper_trip_temp);
+	if (ret)
+		goto unlock;
+
+	if (tz->ops->get_trip_hyst)
+		tz->ops->get_trip_hyst(tz, trip + 1, &upper_trip_hyst);
+
+	upper_trip_temp -= upper_trip_hyst;
+	if (tz->last_temperature > upper_trip_temp &&
+	    upper_trip_temp > tz->temperature)
+		kobject_uevent_env(&tz->device.kobj, KOBJ_CHANGE, msg);
+
+unlock:
+	mutex_unlock(&tz->lock);
+}
+
 static void handle_critical_trips(struct thermal_zone_device *tz,
 				int trip, enum thermal_trip_type trip_type)
 {
@@ -430,6 +480,8 @@ static void handle_critical_trips(struct thermal_zone_device *tz,
 
 	tz->ops->get_trip_temp(tz, trip, &trip_temp);
 
+	thermal_tripped_notify(tz, trip, trip_type, trip_temp);
+
 	/* If we have not crossed the trip_temp, we do not care. */
 	if (trip_temp <= 0 || tz->temperature < trip_temp)
 		return;
-- 
2.5.0

[PATCHv2 0/3] thermal: rework core to improve userspace interaction

[Eduardo Valentin]

Hello Rui, linux-pm

Changelog:
V1 -> V2: Fixes patch 2, and added Srivinas reviewed by on patch 1.

Please consider these three patches in the thermal core to improve
the interaction with userspace.

The first is already in its second version. It avoids reconfiguring
monitor period. Now the thermal core configures the monitor only
after handling all trip points.

The second is about improving emul_temp. The background here is to
allow using emul_temp, even if the thermal zone is not fully setup,
with a missing .get_temp().

The third is to improve hot trip points handling. Hot trip points
are described as notification entry points. However, we do very
little on them. This patch adds a uevent to propagate the event
to userspace. Today, we rely on thermal zone driver. I believe
having the same message coming on every thermal zone makes more
sense. However, I did not remove the .notify() callback, and it
should behave the same.

Eduardo Valentin (3):
  thermal: setup monitor only once after handling trips
  thermal: rework core to allow emul_temp to be treated as regular temp
  thermal: improve hot trip handling

 drivers/thermal/thermal_core.c | 96 ++++++++++++++++++++++++++++++++++--------
 1 file changed, 78 insertions(+), 18 deletions(-)

-- 
2.5.0

[PATCHv2 1/3] thermal: setup monitor only once after handling trips

[Eduardo Valentin]

Instead of changing the monitoring setup every time after
handling each trip, this patch simplifies the monitoring
setup by moving the setup call to a place where all
trips have been treated already.

Cc: Zhang Rui <rui.zhang@intel.com>
Cc: linux-pm@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Reviewed-by: Pandruvada, Srinivas <srinivas.pandruvada@intel.com>
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
---
V2: Added Srinivas reviewed by.
---
 drivers/thermal/thermal_core.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/thermal/thermal_core.c b/drivers/thermal/thermal_core.c
index d9e525c..8fa82c0 100644
--- a/drivers/thermal/thermal_core.c
+++ b/drivers/thermal/thermal_core.c
@@ -457,11 +457,6 @@ static void handle_thermal_trip(struct thermal_zone_device *tz, int trip)
 		handle_critical_trips(tz, trip, type);
 	else
 		handle_non_critical_trips(tz, trip, type);
-	/*
-	 * Alright, we handled this trip successfully.
-	 * So, start monitoring again.
-	 */
-	monitor_thermal_zone(tz);
 }
 
 /**
@@ -547,6 +542,12 @@ void thermal_zone_device_update(struct thermal_zone_device *tz)
 
 	for (count = 0; count < tz->trips; count++)
 		handle_thermal_trip(tz, count);
+
+	/*
+	 * Alright, we handled these trips successfully.
+	 * So, start monitoring again.
+	 */
+	monitor_thermal_zone(tz);
 }
 EXPORT_SYMBOL_GPL(thermal_zone_device_update);
 
-- 
2.5.0