Skip to contentSkip to conversation

thermal/drivers/mediatek/lvts_thermal: Fixes to the interrupt handling

PATCH Applied Nícolas F. R. A. Prado 4 participants 18 messages Jul 6, 2023 – Jul 13, 2023

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

v3 cover thermal/drivers/mediatek/lvts_thermal: Fixes to the interrupt handling

This series fixes interrupt handling in the LVTS thermal driver. It not only solves the interrupt storms currently happening, but also allows the temperature monitoring interrupts to trigger when the thermal trip temperature is reached.

This series was tested together with the MT8192 support series [1] on the Spherion Chromebook (mt8192-asurada-spherion).

These are standalone fixes and don't depend on anything else.

[1] https://lore.kernel.org/all/[email protected]

Thanks,
Nícolas

Changes in v3:
- Reworded cover letter
- Split bitmaps for immediate and filtered modes into separate arrays (patch 2)
- Removed duplicate code for setting MINIMUM_THRESHOLD (patch 5)

Changes in v2:
- Added commits 3, 5, 6 to get working interrupts when crossing thermal trip points
- Updated commit 4 with interrupt flags for the offset

Nícolas F. R. A. Prado (6):
thermal/drivers/mediatek/lvts_thermal: Handle IRQ on all controllers thermal/drivers/mediatek/lvts_thermal: Honor sensors in immediate mode thermal/drivers/mediatek/lvts_thermal: Use offset threshold for IRQ thermal/drivers/mediatek/lvts_thermal: Disable undesired interrupts thermal/drivers/mediatek/lvts_thermal: Don't leave threshold zeroed thermal/drivers/mediatek/lvts_thermal: Manage threshold between
sensors

 drivers/thermal/mediatek/lvts_thermal.c | 152 ++++++++++++++++++------
 1 file changed, 115 insertions(+), 37 deletions(-)
Signature · 1 line
2.41.0
#1 lore raw

PATCH v3 1/6 thermal/drivers/mediatek/lvts_thermal: Handle IRQ on all controllers

There is a single IRQ handler for each LVTS thermal domain, and it is supposed to check each of its underlying controllers for the origin of the interrupt and clear its status. However due to a typo, only the first controller was ever being handled, which resulted in the interrupt never being cleared when it happened on the other controllers. Add the missing index so interrupts are handled for all controllers.

Fixes f5f633b18234 ("thermal/drivers/mediatek: Add the Low Voltage Thermal Sensor driver")

---

(no changes since v1)

 drivers/thermal/mediatek/lvts_thermal.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/thermal/mediatek/lvts_thermal.c +1 -1
index 1e11defe4f35..ba8f86ee12b6 100644
@@ -451,7 +451,7 @@ static irqreturn_t lvts_irq_handler(int irq, void *data)
 
 	for (i = 0; i < lvts_td->num_lvts_ctrl; i++) {
 
-		aux = lvts_ctrl_irq_handler(lvts_td->lvts_ctrl);
+		aux = lvts_ctrl_irq_handler(&lvts_td->lvts_ctrl[i]);
 		if (aux != IRQ_HANDLED)
 			continue;
 
Signature · 1 line
2.41.0
Reviewed-by on v3 1/6
#2 lore raw

Re: [PATCH v3 1/6] thermal/drivers/mediatek/lvts_thermal: Handle IRQ on all controllers

Quoted 6 lines · Nícolas F. R. A. Prado

There is a single IRQ handler for each LVTS thermal domain, and it is supposed to check each of its underlying controllers for the origin of the interrupt and clear its status. However due to a typo, only the first controller was ever being handled, which resulted in the interrupt never being cleared when it happened on the other controllers. Add the missing index so interrupts are handled for all controllers.

Signature · 2 lines
Regards,
Alexandre
#3 lore raw

PATCH v3 2/6 thermal/drivers/mediatek/lvts_thermal: Honor sensors in immediate mode

In reply to Nícolas F. R. A. Prado

Each controller can be configured to operate on immediate or filtered mode. On filtered mode, the sensors are enabled by setting the corresponding bits in MONCTL0, while on immediate mode, by setting MSRCTL1.

Previously, the code would set MSRCTL1 for all four sensors when configured to immediate mode, but given that the controller might not have all four sensors connected, this would cause interrupts to trigger for non-existent sensors. Fix this by handling the MSRCTL1 register analogously to the MONCTL0: only enable the sensors that were declared.

Fixes f5f633b18234 ("thermal/drivers/mediatek: Add the Low Voltage Thermal Sensor driver")

---

Changes in v3:
- Split bitmaps for immediate and filtered modes into separate arrays

 drivers/thermal/mediatek/lvts_thermal.c | 57 ++++++++++++++-----------
 1 file changed, 33 insertions(+), 24 deletions(-)
drivers/thermal/mediatek/lvts_thermal.c +33 -24
index ba8f86ee12b6..8082195f53ae 100644
@@ -896,24 +896,6 @@ static int lvts_ctrl_configure(struct device *dev, struct lvts_ctrl *lvts_ctrl)
 			LVTS_HW_FILTER << 3 | LVTS_HW_FILTER;
 	writel(value, LVTS_MSRCTL0(lvts_ctrl->base));
 
-	/*
-	 * LVTS_MSRCTL1 : Measurement control
-	 *
-	 * Bits:
-	 *
-	 * 9: Ignore MSRCTL0 config and do immediate measurement on sensor3
-	 * 6: Ignore MSRCTL0 config and do immediate measurement on sensor2
-	 * 5: Ignore MSRCTL0 config and do immediate measurement on sensor1
-	 * 4: Ignore MSRCTL0 config and do immediate measurement on sensor0
-	 *
-	 * That configuration will ignore the filtering and the delays
-	 * introduced below in MONCTL1 and MONCTL2
-	 */
-	if (lvts_ctrl->mode == LVTS_MSR_IMMEDIATE_MODE) {
-		value = BIT(9) | BIT(6) | BIT(5) | BIT(4);
-		writel(value, LVTS_MSRCTL1(lvts_ctrl->base));
-	}
-
 	/*
 	 * LVTS_MONCTL1 : Period unit and group interval configuration
 	 *
@@ -979,6 +961,15 @@ static int lvts_ctrl_start(struct device *dev, struct lvts_ctrl *lvts_ctrl)
 	struct thermal_zone_device *tz;
 	u32 sensor_map = 0;
 	int i;
+	/*
+	 * Bitmaps to enable each sensor on immediate and filtered modes, as
+	 * described in MSRCTL1 and MONCTL0 registers below, respectively.
+	 */
+	u32 sensor_imm_bitmap[] = { BIT(4), BIT(5), BIT(6), BIT(9) };
+	u32 sensor_filt_bitmap[] = { BIT(0), BIT(1), BIT(2), BIT(3) };
+
+	u32 *sensor_bitmap = lvts_ctrl->mode == LVTS_MSR_IMMEDIATE_MODE ?
+			     sensor_imm_bitmap : sensor_filt_bitmap;
 
 	for (i = 0; i < lvts_ctrl->num_lvts_sensor; i++) {
 
@@ -1017,20 +1008,38 @@ static int lvts_ctrl_start(struct device *dev, struct lvts_ctrl *lvts_ctrl)
 		 * map, so we can enable the temperature monitoring in
 		 * the hardware thermal controller.
 		 */
-		sensor_map |= BIT(i);
+		sensor_map |= sensor_bitmap[i];
 	}
 
 	/*
-	 * Bits:
-	 *      9: Single point access flow
-	 *    0-3: Enable sensing point 0-3
-	 *
 	 * The initialization of the thermal zones give us
 	 * which sensor point to enable. If any thermal zone
 	 * was not described in the device tree, it won't be
 	 * enabled here in the sensor map.
 	 */
-	writel(sensor_map | BIT(9), LVTS_MONCTL0(lvts_ctrl->base));
+	if (lvts_ctrl->mode == LVTS_MSR_IMMEDIATE_MODE) {
+		/*
+		 * LVTS_MSRCTL1 : Measurement control
+		 *
+		 * Bits:
+		 *
+		 * 9: Ignore MSRCTL0 config and do immediate measurement on sensor3
+		 * 6: Ignore MSRCTL0 config and do immediate measurement on sensor2
+		 * 5: Ignore MSRCTL0 config and do immediate measurement on sensor1
+		 * 4: Ignore MSRCTL0 config and do immediate measurement on sensor0
+		 *
+		 * That configuration will ignore the filtering and the delays
+		 * introduced in MONCTL1 and MONCTL2
+		 */
+		writel(sensor_map, LVTS_MSRCTL1(lvts_ctrl->base));
+	} else {
+		/*
+		 * Bits:
+		 *      9: Single point access flow
+		 *    0-3: Enable sensing point 0-3
+		 */
+		writel(sensor_map | BIT(9), LVTS_MONCTL0(lvts_ctrl->base));
+	}
 
 	return 0;
 }
Signature · 1 line
2.41.0
Reviewed-by on v3 2/6
#4 lore raw

Re: [PATCH v3 2/6] thermal/drivers/mediatek/lvts_thermal: Honor sensors in immediate mode

Quoted 10 lines · Nícolas F. R. A. Prado

Each controller can be configured to operate on immediate or filtered mode. On filtered mode, the sensors are enabled by setting the corresponding bits in MONCTL0, while on immediate mode, by setting MSRCTL1.

Previously, the code would set MSRCTL1 for all four sensors when configured to immediate mode, but given that the controller might not have all four sensors connected, this would cause interrupts to trigger for non-existent sensors. Fix this by handling the MSRCTL1 register analogously to the MONCTL0: only enable the sensors that were declared.

Signature · 2 lines
Regards,
Alexandre
#5 lore raw

PATCH v3 3/6 thermal/drivers/mediatek/lvts_thermal: Use offset threshold for IRQ

In reply to Nícolas F. R. A. Prado

There are two kinds of temperature monitoring interrupts available:
* High Offset, Low Offset
* Hot, Hot to normal, Cold

The code currently uses the hot/h2n/cold interrupts, however in a way that doesn't work: the cold threshold is left uninitialized, which prevents the other thresholds from ever triggering, and the h2n interrupt is used as the lower threshold, which prevents the hot interrupt from triggering again after the thresholds are updated by the thermal framework, since a hot interrupt can only trigger again after the hot to normal interrupt has been triggered.

But better yet than addressing those issues, is to use the high/low offset interrupts instead. This way only two thresholds need to be managed, which have a simpler state machine, making them a better match to the thermal framework's high and low thresholds.

Fixes f5f633b18234 ("thermal/drivers/mediatek: Add the Low Voltage Thermal Sensor driver")

---

(no changes since v2)

Changes in v2:
- Added this commit

 drivers/thermal/mediatek/lvts_thermal.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
drivers/thermal/mediatek/lvts_thermal.c +6 -6
index 8082195f53ae..e7cbfe0426b5 100644
@@ -298,9 +298,9 @@ static int lvts_set_trips(struct thermal_zone_device *tz, int low, int high)
 	u32 raw_high = lvts_temp_to_raw(high);
 
 	/*
-	 * Hot to normal temperature threshold
+	 * Low offset temperature threshold
 	 *
-	 * LVTS_H2NTHRE
+	 * LVTS_OFFSETL
 	 *
 	 * Bits:
 	 *
@@ -309,13 +309,13 @@ static int lvts_set_trips(struct thermal_zone_device *tz, int low, int high)
 	if (low != -INT_MAX) {
 		pr_debug("%s: Setting low limit temperature interrupt: %d\n",
 			 thermal_zone_device_type(tz), low);
-		writel(raw_low, LVTS_H2NTHRE(base));
+		writel(raw_low, LVTS_OFFSETL(base));
 	}
 
 	/*
-	 * Hot temperature threshold
+	 * High offset temperature threshold
 	 *
-	 * LVTS_HTHRE
+	 * LVTS_OFFSETH
 	 *
 	 * Bits:
 	 *
@@ -323,7 +323,7 @@ static int lvts_set_trips(struct thermal_zone_device *tz, int low, int high)
 	 */
 	pr_debug("%s: Setting high limit temperature interrupt: %d\n",
 		 thermal_zone_device_type(tz), high);
-	writel(raw_high, LVTS_HTHRE(base));
+	writel(raw_high, LVTS_OFFSETH(base));
 
 	return 0;
 }
Signature · 1 line
2.41.0
maintainer Reviewed-by on v3 3/6
#6 lore raw

Re: [PATCH v3 3/6] thermal/drivers/mediatek/lvts_thermal: Use offset threshold for IRQ

Il 06/07/23 17:37, Nícolas F. R. A. Prado ha scritto:

Quoted 12 lines

There are two kinds of temperature monitoring interrupts available:
* High Offset, Low Offset
* Hot, Hot to normal, Cold

The code currently uses the hot/h2n/cold interrupts, however in a way that doesn't work: the cold threshold is left uninitialized, which prevents the other thresholds from ever triggering, and the h2n interrupt is used as the lower threshold, which prevents the hot interrupt from triggering again after the thresholds are updated by the thermal framework, since a hot interrupt can only trigger again after the hot to normal interrupt has been triggered.

Quoted 8 lines

But better yet than addressing those issues, is to use the high/low offset interrupts instead. This way only two thresholds need to be managed, which have a simpler state machine, making them a better match to the thermal framework's high and low thresholds.

Fixes f5f633b18234 ("thermal/drivers/mediatek: Add the Low Voltage Thermal Sensor driver")
Reviewed-by on v3 3/6
#7 lore raw

Re: [PATCH v3 3/6] thermal/drivers/mediatek/lvts_thermal: Use offset threshold for IRQ

In reply to Nícolas F. R. A. Prado
Quoted 12 lines · Nícolas F. R. A. Prado

There are two kinds of temperature monitoring interrupts available:
* High Offset, Low Offset
* Hot, Hot to normal, Cold

The code currently uses the hot/h2n/cold interrupts, however in a way that doesn't work: the cold threshold is left uninitialized, which prevents the other thresholds from ever triggering, and the h2n interrupt is used as the lower threshold, which prevents the hot interrupt from triggering again after the thresholds are updated by the thermal framework, since a hot interrupt can only trigger again after the hot to normal interrupt has been triggered.

Quoted 4 lines

But better yet than addressing those issues, is to use the high/low offset interrupts instead. This way only two thresholds need to be managed, which have a simpler state machine, making them a better match to the thermal framework's high and low thresholds.

Signature · 2 lines
Regards,
Alexandre
#8 lore raw

PATCH v3 4/6 thermal/drivers/mediatek/lvts_thermal: Disable undesired interrupts

In reply to Nícolas F. R. A. Prado

Out of the many interrupts supported by the hardware, the only ones of interest to the driver currently are:
* The temperature went over the high offset threshold, for any of the sensors
* The temperature went below the low offset threshold, for any of the sensors
* The temperature went over the stage3 threshold

These are the only thresholds configured by the driver through the OFFSETH, OFFSETL, and PROTTC registers, respectively.

The current interrupt mask in LVTS_MONINT_CONF, enables many more interrupts, including data ready on sensors for both filtered and immediate mode. These are not only not handled by the driver, but they are also triggered too often, causing unneeded overhead. Disable these unnecessary interrupts.

The meaning of each bit can be seen in the comment describing LVTS_MONINTST in the IRQ handler.

Fixes f5f633b18234 ("thermal/drivers/mediatek: Add the Low Voltage Thermal Sensor driver")

---

(no changes since v2)

Changes in v2:
- Reworded commit and changed flag to use offset interrupts instead

 drivers/thermal/mediatek/lvts_thermal.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/thermal/mediatek/lvts_thermal.c +1 -1
index e7cbfe0426b5..e6dd4d120e54 100644
@@ -65,7 +65,7 @@
 #define LVTS_HW_FILTER				0x2
 #define LVTS_TSSEL_CONF				0x13121110
 #define LVTS_CALSCALE_CONF			0x300
-#define LVTS_MONINT_CONF			0x9FBF7BDE
+#define LVTS_MONINT_CONF			0x8300318C
 
 #define LVTS_INT_SENSOR0			0x0009001F
 #define LVTS_INT_SENSOR1			0x001203E0
Signature · 1 line
2.41.0
maintainer Reviewed-by on v3 4/6
#9 lore raw

Re: [PATCH v3 4/6] thermal/drivers/mediatek/lvts_thermal: Disable undesired interrupts

Il 06/07/23 17:37, Nícolas F. R. A. Prado ha scritto:

Quoted 17 lines

Out of the many interrupts supported by the hardware, the only ones of interest to the driver currently are:
* The temperature went over the high offset threshold, for any of the sensors
* The temperature went below the low offset threshold, for any of the sensors
* The temperature went over the stage3 threshold

These are the only thresholds configured by the driver through the OFFSETH, OFFSETL, and PROTTC registers, respectively.

The current interrupt mask in LVTS_MONINT_CONF, enables many more interrupts, including data ready on sensors for both filtered and immediate mode. These are not only not handled by the driver, but they are also triggered too often, causing unneeded overhead. Disable these unnecessary interrupts.

Quoted 6 lines

The meaning of each bit can be seen in the comment describing LVTS_MONINTST in the IRQ handler.

Fixes f5f633b18234 ("thermal/drivers/mediatek: Add the Low Voltage Thermal Sensor driver")
Reviewed-by on v3 4/6
#10 lore raw

Re: [PATCH v3 4/6] thermal/drivers/mediatek/lvts_thermal: Disable undesired interrupts

In reply to Nícolas F. R. A. Prado
Quoted 11 lines · Nícolas F. R. A. Prado

Out of the many interrupts supported by the hardware, the only ones of interest to the driver currently are:
* The temperature went over the high offset threshold, for any of the sensors
* The temperature went below the low offset threshold, for any of the sensors
* The temperature went over the stage3 threshold

These are the only thresholds configured by the driver through the OFFSETH, OFFSETL, and PROTTC registers, respectively.

Quoted 8 lines

The current interrupt mask in LVTS_MONINT_CONF, enables many more interrupts, including data ready on sensors for both filtered and immediate mode. These are not only not handled by the driver, but they are also triggered too often, causing unneeded overhead. Disable these unnecessary interrupts.

The meaning of each bit can be seen in the comment describing LVTS_MONINTST in the IRQ handler.

Signature · 2 lines
Regards,
Alexandre
#11 lore raw

PATCH v3 5/6 thermal/drivers/mediatek/lvts_thermal: Don't leave threshold zeroed

In reply to Nícolas F. R. A. Prado

The thermal framework might leave the low threshold unset if there aren't any lower trip points. This leaves the register zeroed, which translates to a very high temperature for the low threshold. The interrupt for this threshold is then immediately triggered, and the state machine gets stuck, preventing any other temperature monitoring interrupts to ever trigger.

(The same happens by not setting the Cold or Hot to Normal thresholds when using those)

Set the unused threshold to a valid low value. This value was chosen so that for any valid golden temperature read from the efuse, when the value is converted to raw and back again to milliCelsius, the result doesn't underflow.

Fixes f5f633b18234 ("thermal/drivers/mediatek: Add the Low Voltage Thermal Sensor driver")

---

Changes in v3:
- Converted MINIMUM_THRESHOLD earlier to eliminate need for extra branch
- Fixed typo: MINIUM -> MINIMUM

Changes in v2:
- Added this commit

 drivers/thermal/mediatek/lvts_thermal.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
drivers/thermal/mediatek/lvts_thermal.c +6 -6
index e6dd4d120e54..7552e1d59dc9 100644
@@ -83,6 +83,8 @@
 
 #define LVTS_HW_SHUTDOWN_MT8195		105000
 
+#define LVTS_MINIMUM_THRESHOLD		20000
+
 static int golden_temp = LVTS_GOLDEN_TEMP_DEFAULT;
 static int coeff_b = LVTS_COEFF_B;
 
@@ -294,7 +296,7 @@ static int lvts_set_trips(struct thermal_zone_device *tz, int low, int high)
 {
 	struct lvts_sensor *lvts_sensor = thermal_zone_device_priv(tz);
 	void __iomem *base = lvts_sensor->base;
-	u32 raw_low = lvts_temp_to_raw(low);
+	u32 raw_low = lvts_temp_to_raw(low != -INT_MAX ? low : LVTS_MINIMUM_THRESHOLD);
 	u32 raw_high = lvts_temp_to_raw(high);
 
 	/*
@@ -306,11 +308,9 @@ static int lvts_set_trips(struct thermal_zone_device *tz, int low, int high)
 	 *
 	 * 14-0 : Raw temperature for threshold
 	 */
-	if (low != -INT_MAX) {
-		pr_debug("%s: Setting low limit temperature interrupt: %d\n",
-			 thermal_zone_device_type(tz), low);
-		writel(raw_low, LVTS_OFFSETL(base));
-	}
+	pr_debug("%s: Setting low limit temperature interrupt: %d\n",
+		 thermal_zone_device_type(tz), low);
+	writel(raw_low, LVTS_OFFSETL(base));
 
 	/*
 	 * High offset temperature threshold
Signature · 1 line
2.41.0
maintainer Reviewed-by on v3 5/6
#12 lore raw

Re: [PATCH v3 5/6] thermal/drivers/mediatek/lvts_thermal: Don't leave threshold zeroed

Il 06/07/23 17:37, Nícolas F. R. A. Prado ha scritto:

Quoted 10 lines

The thermal framework might leave the low threshold unset if there aren't any lower trip points. This leaves the register zeroed, which translates to a very high temperature for the low threshold. The interrupt for this threshold is then immediately triggered, and the state machine gets stuck, preventing any other temperature monitoring interrupts to ever trigger.

(The same happens by not setting the Cold or Hot to Normal thresholds when using those)

Quoted 8 lines

Set the unused threshold to a valid low value. This value was chosen so that for any valid golden temperature read from the efuse, when the value is converted to raw and back again to milliCelsius, the result doesn't underflow.

Fixes f5f633b18234 ("thermal/drivers/mediatek: Add the Low Voltage Thermal Sensor driver")
Reviewed-by on v3 5/6
#13 lore raw

Re: [PATCH v3 5/6] thermal/drivers/mediatek/lvts_thermal: Don't leave threshold zeroed

In reply to Nícolas F. R. A. Prado
Quoted 7 lines · Nícolas F. R. A. Prado

The thermal framework might leave the low threshold unset if there aren't any lower trip points. This leaves the register zeroed, which translates to a very high temperature for the low threshold. The interrupt for this threshold is then immediately triggered, and the state machine gets stuck, preventing any other temperature monitoring interrupts to ever trigger.

Quoted 7 lines

(The same happens by not setting the Cold or Hot to Normal thresholds when using those)

Set the unused threshold to a valid low value. This value was chosen so that for any valid golden temperature read from the efuse, when the value is converted to raw and back again to milliCelsius, the result doesn't underflow.

Signature · 2 lines
Regards,
Alexandre
#14 lore raw

PATCH v3 6/6 thermal/drivers/mediatek/lvts_thermal: Manage threshold between sensors

In reply to Nícolas F. R. A. Prado

Each LVTS thermal controller can have up to four sensors, each capable of triggering its own interrupt when its measured temperature crosses the configured threshold. The threshold for each sensor is handled separately by the thermal framework, since each one is registered with its own thermal zone and trips. However, the temperature thresholds are configured on the controller, and therefore are shared between all sensors on that controller.

When the temperature measured by the sensors is different enough to cause the thermal framework to configure different thresholds for each one, interrupts start triggering on sensors outside the last threshold configured.

To address the issue, track the thresholds required by each sensor and only actually set the highest one in the hardware, and disable interrupts for all sensors outside the current configured range.

Fixes f5f633b18234 ("thermal/drivers/mediatek: Add the Low Voltage Thermal Sensor driver")

---

(no changes since v2)

Changes in v2:
- Added this commit

 drivers/thermal/mediatek/lvts_thermal.c | 69 +++++++++++++++++++++++++
 1 file changed, 69 insertions(+)
drivers/thermal/mediatek/lvts_thermal.c +69 -0
index 7552e1d59dc9..a2dc33697371 100644
@@ -67,6 +67,11 @@
 #define LVTS_CALSCALE_CONF			0x300
 #define LVTS_MONINT_CONF			0x8300318C
 
+#define LVTS_MONINT_OFFSET_SENSOR0		0xC
+#define LVTS_MONINT_OFFSET_SENSOR1		0x180
+#define LVTS_MONINT_OFFSET_SENSOR2		0x3000
+#define LVTS_MONINT_OFFSET_SENSOR3		0x3000000
+
 #define LVTS_INT_SENSOR0			0x0009001F
 #define LVTS_INT_SENSOR1			0x001203E0
 #define LVTS_INT_SENSOR2			0x00247C00
@@ -112,6 +117,8 @@ struct lvts_sensor {
 	void __iomem *base;
 	int id;
 	int dt_id;
+	int low_thresh;
+	int high_thresh;
 };
 
 struct lvts_ctrl {
@@ -121,6 +128,8 @@ struct lvts_ctrl {
 	int num_lvts_sensor;
 	int mode;
 	void __iomem *base;
+	int low_thresh;
+	int high_thresh;
 };
 
 struct lvts_domain {
@@ -292,12 +301,66 @@ static int lvts_get_temp(struct thermal_zone_device *tz, int *temp)
 	return 0;
 }
 
+static void lvts_update_irq_mask(struct lvts_ctrl *lvts_ctrl)
+{
+	u32 masks[] = {
+		LVTS_MONINT_OFFSET_SENSOR0,
+		LVTS_MONINT_OFFSET_SENSOR1,
+		LVTS_MONINT_OFFSET_SENSOR2,
+		LVTS_MONINT_OFFSET_SENSOR3,
+	};
+	u32 value = 0;
+	int i;
+
+	value = readl(LVTS_MONINT(lvts_ctrl->base));
+
+	for (i = 0; i < ARRAY_SIZE(masks); i++) {
+		if (lvts_ctrl->sensors[i].high_thresh == lvts_ctrl->high_thresh
+		    && lvts_ctrl->sensors[i].low_thresh == lvts_ctrl->low_thresh)
+			value |= masks[i];
+		else
+			value &= ~masks[i];
+	}
+
+	writel(value, LVTS_MONINT(lvts_ctrl->base));
+}
+
+static bool lvts_should_update_thresh(struct lvts_ctrl *lvts_ctrl, int high)
+{
+	int i;
+
+	if (high > lvts_ctrl->high_thresh)
+		return true;
+
+	for (i = 0; i < lvts_ctrl->num_lvts_sensor; i++)
+		if (lvts_ctrl->sensors[i].high_thresh == lvts_ctrl->high_thresh
+		    && lvts_ctrl->sensors[i].low_thresh == lvts_ctrl->low_thresh)
+			return false;
+
+	return true;
+}
+
 static int lvts_set_trips(struct thermal_zone_device *tz, int low, int high)
 {
 	struct lvts_sensor *lvts_sensor = thermal_zone_device_priv(tz);
+	struct lvts_ctrl *lvts_ctrl = container_of(lvts_sensor, struct lvts_ctrl, sensors[lvts_sensor->id]);
 	void __iomem *base = lvts_sensor->base;
 	u32 raw_low = lvts_temp_to_raw(low != -INT_MAX ? low : LVTS_MINIMUM_THRESHOLD);
 	u32 raw_high = lvts_temp_to_raw(high);
+	bool should_update_thresh;
+
+	lvts_sensor->low_thresh = low;
+	lvts_sensor->high_thresh = high;
+
+	should_update_thresh = lvts_should_update_thresh(lvts_ctrl, high);
+	if (should_update_thresh) {
+		lvts_ctrl->high_thresh = high;
+		lvts_ctrl->low_thresh = low;
+	}
+	lvts_update_irq_mask(lvts_ctrl);
+
+	if (!should_update_thresh)
+		return 0;
 
 	/*
 	 * Low offset temperature threshold
@@ -521,6 +584,9 @@ static int lvts_sensor_init(struct device *dev, struct lvts_ctrl *lvts_ctrl,
 		 */
 		lvts_sensor[i].msr = lvts_ctrl_data->mode == LVTS_MSR_IMMEDIATE_MODE ?
 			imm_regs[i] : msr_regs[i];
+
+		lvts_sensor[i].low_thresh = INT_MIN;
+		lvts_sensor[i].high_thresh = INT_MIN;
 	};
 
 	lvts_ctrl->num_lvts_sensor = lvts_ctrl_data->num_lvts_sensor;
@@ -688,6 +754,9 @@ static int lvts_ctrl_init(struct device *dev, struct lvts_domain *lvts_td,
 		 */
 		lvts_ctrl[i].hw_tshut_raw_temp =
 			lvts_temp_to_raw(lvts_data->lvts_ctrl[i].hw_tshut_temp);
+
+		lvts_ctrl[i].low_thresh = INT_MIN;
+		lvts_ctrl[i].high_thresh = INT_MIN;
 	}
 
 	/*
Signature · 1 line
2.41.0
maintainer Reviewed-by on v3 6/6
#15 lore raw

Re: [PATCH v3 6/6] thermal/drivers/mediatek/lvts_thermal: Manage threshold between sensors

Il 06/07/23 17:37, Nícolas F. R. A. Prado ha scritto:

Quoted 13 lines

Each LVTS thermal controller can have up to four sensors, each capable of triggering its own interrupt when its measured temperature crosses the configured threshold. The threshold for each sensor is handled separately by the thermal framework, since each one is registered with its own thermal zone and trips. However, the temperature thresholds are configured on the controller, and therefore are shared between all sensors on that controller.

When the temperature measured by the sensors is different enough to cause the thermal framework to configure different thresholds for each one, interrupts start triggering on sensors outside the last threshold configured.

Quoted 7 lines

To address the issue, track the thresholds required by each sensor and only actually set the highest one in the hardware, and disable interrupts for all sensors outside the current configured range.

Fixes f5f633b18234 ("thermal/drivers/mediatek: Add the Low Voltage Thermal Sensor driver")
Reviewed-by on v3 6/6
#16 lore raw

Re: [PATCH v3 6/6] thermal/drivers/mediatek/lvts_thermal: Manage threshold between sensors

In reply to Nícolas F. R. A. Prado
Quoted 8 lines · Nícolas F. R. A. Prado

Each LVTS thermal controller can have up to four sensors, each capable of triggering its own interrupt when its measured temperature crosses the configured threshold. The threshold for each sensor is handled separately by the thermal framework, since each one is registered with its own thermal zone and trips. However, the temperature thresholds are configured on the controller, and therefore are shared between all sensors on that controller.

Quoted 8 lines

When the temperature measured by the sensors is different enough to cause the thermal framework to configure different thresholds for each one, interrupts start triggering on sensors outside the last threshold configured.

To address the issue, track the thresholds required by each sensor and only actually set the highest one in the hardware, and disable interrupts for all sensors outside the current configured range.

Signature · 2 lines
Regards,
Alexandre
#17 lore raw
In reply to Nícolas F. R. A. Prado
Quoted 6 lines · Nícolas F. R. A. Prado

This series fixes interrupt handling in the LVTS thermal driver. It not only solves the interrupt storms currently happening, but also allows the temperature monitoring interrupts to trigger when the thermal trip temperature is reached.

Quoted 6 lines

This series was tested together with the MT8192 support series [1] on the Spherion Chromebook (mt8192-asurada-spherion).

These are standalone fixes and don't depend on anything else.

[1] https://lore.kernel.org/all/[email protected]

Applied, thanks

Signature · 5 lines
<http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs

Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog