linux-pm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH v5 0/9] qcom: pm8150: add support for thermal monitoring
@ 2020-09-14 15:48 Dmitry Baryshkov
  2020-09-14 15:48 ` [PATCH v5 1/9] dt-bindings: thermal: qcom: add adc-thermal monitor bindings Dmitry Baryshkov
                   ` (8 more replies)
  0 siblings, 9 replies; 15+ messages in thread
From: Dmitry Baryshkov @ 2020-09-14 15:48 UTC (permalink / raw)
  To: Andy Gross, Bjorn Andersson, Rob Herring, Zhang Rui,
	Daniel Lezcano, Amit Kucheria, Jonathan Cameron, Hartmut Knaack,
	Lars-Peter Clausen, Peter Meerwald-Stadler
  Cc: linux-arm-msm, linux-pm, devicetree, linux-iio, Manivannan Sadhasivam

This patch serie adds support for thermal monitoring block on Qualcomm's
PMIC5 chips. PM8150{,b,l} and sm8250-mtp board device trees are extended
to support thermal zones provided by this thermal monitoring block.
Unlike the rest of PMIC thermal senses, these thermal zones describe
particular thermistors, which differ between from board to board.

Changes since v4:
 - Added kernel-doc comments to ADC-TM structures
 - Used several sizeof(buf) instead of hand-conding register size

Changes since v3:
 - Fix DT description to spell "thermal monitoring" instead of just TM
 - Fix warnings in DT example
 - Add EXPORT_SYMBOL_GPL(of_iio_channel_get_by_name)
 - Fixed whitespace chanes in qcom-vadc-common.c
 - Removed error message if IIO chanel get returns -EPROBE_DEFER

Changes since v2:
 - IIO: export of_iio_channel_get_by_name() function
 - dt-bindings: move individual io-channels to each thermal monitoring
   channel rather than listing them all in device node
 - added fallback defaults to of_device_get_match_data calls in
   qcom-spmi-adc5 and qcom-spmi-adc-tm5 drivers
 - minor typo fixes

Changes since v1:
 - Introduce fixp_linear_interpolate() by Craig Tatlor
 - Lots of syntax/whitespace changes
 - Cleaned up register definitions per Jonathan's suggestion
 - Implemented most of the suggestions from Bjorn's and Jonathan's
   review




^ permalink raw reply	[flat|nested] 15+ messages in thread

* [PATCH v5 1/9] dt-bindings: thermal: qcom: add adc-thermal monitor bindings
  2020-09-14 15:48 [PATCH v5 0/9] qcom: pm8150: add support for thermal monitoring Dmitry Baryshkov
@ 2020-09-14 15:48 ` Dmitry Baryshkov
  2020-09-22 23:40   ` Rob Herring
  2020-09-14 15:48 ` [PATCH v5 2/9] fixp-arith: add a linear interpolation function Dmitry Baryshkov
                   ` (7 subsequent siblings)
  8 siblings, 1 reply; 15+ messages in thread
From: Dmitry Baryshkov @ 2020-09-14 15:48 UTC (permalink / raw)
  To: Andy Gross, Bjorn Andersson, Rob Herring, Zhang Rui,
	Daniel Lezcano, Amit Kucheria, Jonathan Cameron, Hartmut Knaack,
	Lars-Peter Clausen, Peter Meerwald-Stadler
  Cc: linux-arm-msm, linux-pm, devicetree, linux-iio,
	Manivannan Sadhasivam, Jonathan Cameron

Add bindings for thermal monitor, part of Qualcomm PMIC5 chips. It is a
close counterpart of VADC part of those PMICs.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
---
 .../bindings/thermal/qcom-spmi-adc-tm5.yaml   | 151 ++++++++++++++++++
 1 file changed, 151 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml

diff --git a/Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml b/Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml
new file mode 100644
index 000000000000..432a65839b89
--- /dev/null
+++ b/Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml
@@ -0,0 +1,151 @@
+# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/thermal/qcom-spmi-adc-tm5.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm's SPMI PMIC ADC Thermal Monitoring
+maintainers:
+  - Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
+
+properties:
+  compatible:
+    const: qcom,spmi-adc-tm5
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    maxItems: 1
+
+  "#thermal-sensor-cells":
+    const: 1
+    description:
+      Number of cells required to uniquely identify the thermal sensors. Since
+      we have multiple sensors this is set to 1
+
+  "#address-cells":
+    const: 1
+
+  "#size-cells":
+    const: 0
+
+  qcom,avg-samples:
+    $ref: /schemas/types.yaml#/definitions/uint32
+    description: Number of samples to be used for measurement.
+    enum:
+      - 1
+      - 2
+      - 4
+      - 8
+      - 16
+    default: 1
+
+  qcom,decimation:
+    $ref: /schemas/types.yaml#/definitions/uint32
+    description: This parameter is used to decrease ADC sampling rate.
+            Quicker measurements can be made by reducing decimation ratio.
+    enum:
+      - 250
+      - 420
+      - 840
+    default: 840
+
+patternProperties:
+  "^([-a-z0-9]*)@[0-9]+$":
+    type: object
+    description:
+      Represent one thermal sensor.
+
+    properties:
+      reg:
+        description: Specify the sensor channel.
+        maxItems: 1
+
+      io-channels:
+        description:
+          From common IIO binding. Used to pipe PMIC ADC channel to thermal monitor
+
+      qcom,adc-channel:
+        $ref: /schemas/types.yaml#/definitions/uint32
+        description: Corresponding ADC channel ID.
+
+      qcom,ratiometric:
+        $ref: /schemas/types.yaml#/definitions/flag
+        description:
+          Channel calibration type.
+          If this property is specified VADC will use the VDD reference
+          (1.875V) and GND for channel calibration. If property is not found,
+          channel will be calibrated with 0V and 1.25V reference channels,
+          also known as absolute calibration.
+
+      qcom,hw-settle-time:
+        $ref: /schemas/types.yaml#/definitions/uint32
+        description: Time between AMUX getting configured and the ADC starting conversion.
+
+      qcom,pre-scaling:
+        $ref: /schemas/types.yaml#/definitions/uint32-array
+        description: Used for scaling the channel input signal before the
+          signal is fed to VADC. See qcom,spi-vadc specification for the list
+          of possible values.
+        minItems: 2
+        maxItems: 2
+
+    required:
+      - reg
+      - qcom,adc-channel
+
+    additionalProperties:
+      false
+
+required:
+  - compatible
+  - reg
+  - interrupts
+  - "#address-cells"
+  - "#size-cells"
+  - "#thermal-sensor-cells"
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/iio/qcom,spmi-vadc.h>
+    #include <dt-bindings/interrupt-controller/irq.h>
+    spmi_bus {
+        #address-cells = <1>;
+        #size-cells = <0>;
+        pm8150b_adc: adc@3100 {
+            reg = <0x3100>;
+            compatible = "qcom,spmi-adc5";
+            #address-cells = <1>;
+            #size-cells = <0>;
+            #io-channel-cells = <1>;
+            io-channel-ranges;
+
+            /* Other propreties are omitted */
+            conn-therm@4f {
+                reg = <ADC5_AMUX_THM3_100K_PU>;
+                qcom,ratiometric;
+                qcom,hw-settle-time = <200>;
+            };
+        };
+
+        pm8150b_adc_tm: adc-tm@3500 {
+            compatible = "qcom,spmi-adc-tm5";
+            reg = <0x3500>;
+            interrupts = <0x2 0x35 0x0 IRQ_TYPE_EDGE_RISING>;
+            #thermal-sensor-cells = <1>;
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            conn-therm@0 {
+                reg = <0>;
+                io-channels = <&pm8150b_adc ADC5_AMUX_THM3_100K_PU>;
+                qcom,adc-channel = <ADC5_AMUX_THM3_100K_PU>;
+                qcom,ratiometric;
+                qcom,hw-settle-time = <200>;
+            };
+        };
+    };
+...
-- 
2.28.0


^ permalink raw reply related	[flat|nested] 15+ messages in thread

* [PATCH v5 2/9] fixp-arith: add a linear interpolation function
  2020-09-14 15:48 [PATCH v5 0/9] qcom: pm8150: add support for thermal monitoring Dmitry Baryshkov
  2020-09-14 15:48 ` [PATCH v5 1/9] dt-bindings: thermal: qcom: add adc-thermal monitor bindings Dmitry Baryshkov
@ 2020-09-14 15:48 ` Dmitry Baryshkov
  2020-09-14 15:48 ` [PATCH v5 3/9] iio: adc: qcom-vadc: move several adc5 functions to common file Dmitry Baryshkov
                   ` (6 subsequent siblings)
  8 siblings, 0 replies; 15+ messages in thread
From: Dmitry Baryshkov @ 2020-09-14 15:48 UTC (permalink / raw)
  To: Andy Gross, Bjorn Andersson, Rob Herring, Zhang Rui,
	Daniel Lezcano, Amit Kucheria, Jonathan Cameron, Hartmut Knaack,
	Lars-Peter Clausen, Peter Meerwald-Stadler
  Cc: linux-arm-msm, linux-pm, devicetree, linux-iio,
	Manivannan Sadhasivam, Craig Tatlor, Jonathan Cameron

From: Craig Tatlor <ctatlor97@gmail.com>

Adds a function to interpolate against two points,
this is carried arount as a helper function by tons of drivers.

Signed-off-by: Craig Tatlor <ctatlor97@gmail.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Acked-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
---
 include/linux/fixp-arith.h | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/include/linux/fixp-arith.h b/include/linux/fixp-arith.h
index 8396013785ef..281cb4f83dbe 100644
--- a/include/linux/fixp-arith.h
+++ b/include/linux/fixp-arith.h
@@ -141,4 +141,23 @@ static inline s32 fixp_sin32_rad(u32 radians, u32 twopi)
 #define fixp_cos32_rad(rad, twopi)	\
 	fixp_sin32_rad(rad + twopi / 4, twopi)
 
+/**
+ * fixp_linear_interpolate() - interpolates a value from two known points
+ *
+ * @x0: x value of point 0
+ * @y0: y value of point 0
+ * @x1: x value of point 1
+ * @y1: y value of point 1
+ * @x: the linear interpolant
+ */
+static inline int fixp_linear_interpolate(int x0, int y0, int x1, int y1, int x)
+{
+	if (y0 == y1 || x == x0)
+		return y0;
+	if (x1 == x0 || x == x1)
+		return y1;
+
+	return y0 + ((y1 - y0) * (x - x0) / (x1 - x0));
+}
+
 #endif
-- 
2.28.0


^ permalink raw reply related	[flat|nested] 15+ messages in thread

* [PATCH v5 3/9] iio: adc: qcom-vadc: move several adc5 functions to common file
  2020-09-14 15:48 [PATCH v5 0/9] qcom: pm8150: add support for thermal monitoring Dmitry Baryshkov
  2020-09-14 15:48 ` [PATCH v5 1/9] dt-bindings: thermal: qcom: add adc-thermal monitor bindings Dmitry Baryshkov
  2020-09-14 15:48 ` [PATCH v5 2/9] fixp-arith: add a linear interpolation function Dmitry Baryshkov
@ 2020-09-14 15:48 ` Dmitry Baryshkov
  2020-09-14 15:48 ` [PATCH v5 4/9] iio: adc: qcom-vadc-common: use fixp_linear_interpolate Dmitry Baryshkov
                   ` (5 subsequent siblings)
  8 siblings, 0 replies; 15+ messages in thread
From: Dmitry Baryshkov @ 2020-09-14 15:48 UTC (permalink / raw)
  To: Andy Gross, Bjorn Andersson, Rob Herring, Zhang Rui,
	Daniel Lezcano, Amit Kucheria, Jonathan Cameron, Hartmut Knaack,
	Lars-Peter Clausen, Peter Meerwald-Stadler
  Cc: linux-arm-msm, linux-pm, devicetree, linux-iio,
	Manivannan Sadhasivam, Jonathan Cameron

ADC-TM5 driver will make use of several functions from ADC5 driver. Move
them to qcom-vadc-common driver.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
---
 drivers/iio/adc/qcom-spmi-adc5.c   | 75 +++---------------------------
 drivers/iio/adc/qcom-vadc-common.c | 68 ++++++++++++++++++++++++++-
 drivers/iio/adc/qcom-vadc-common.h | 10 +++-
 3 files changed, 82 insertions(+), 71 deletions(-)

diff --git a/drivers/iio/adc/qcom-spmi-adc5.c b/drivers/iio/adc/qcom-spmi-adc5.c
index c10aa28be70a..c2da8f068b87 100644
--- a/drivers/iio/adc/qcom-spmi-adc5.c
+++ b/drivers/iio/adc/qcom-spmi-adc5.c
@@ -154,18 +154,6 @@ struct adc5_chip {
 	const struct adc5_data	*data;
 };
 
-static const struct vadc_prescale_ratio adc5_prescale_ratios[] = {
-	{.num =  1, .den =  1},
-	{.num =  1, .den =  3},
-	{.num =  1, .den =  4},
-	{.num =  1, .den =  6},
-	{.num =  1, .den = 20},
-	{.num =  1, .den =  8},
-	{.num = 10, .den = 81},
-	{.num =  1, .den = 10},
-	{.num =  1, .den = 16}
-};
-
 static int adc5_read(struct adc5_chip *adc, u16 offset, u8 *data, int len)
 {
 	return regmap_bulk_read(adc->regmap, adc->base + offset, data, len);
@@ -181,55 +169,6 @@ static int adc5_masked_write(struct adc5_chip *adc, u16 offset, u8 mask, u8 val)
 	return regmap_update_bits(adc->regmap, adc->base + offset, mask, val);
 }
 
-static int adc5_prescaling_from_dt(u32 num, u32 den)
-{
-	unsigned int pre;
-
-	for (pre = 0; pre < ARRAY_SIZE(adc5_prescale_ratios); pre++)
-		if (adc5_prescale_ratios[pre].num == num &&
-		    adc5_prescale_ratios[pre].den == den)
-			break;
-
-	if (pre == ARRAY_SIZE(adc5_prescale_ratios))
-		return -EINVAL;
-
-	return pre;
-}
-
-static int adc5_hw_settle_time_from_dt(u32 value,
-					const unsigned int *hw_settle)
-{
-	unsigned int i;
-
-	for (i = 0; i < VADC_HW_SETTLE_SAMPLES_MAX; i++) {
-		if (value == hw_settle[i])
-			return i;
-	}
-
-	return -EINVAL;
-}
-
-static int adc5_avg_samples_from_dt(u32 value)
-{
-	if (!is_power_of_2(value) || value > ADC5_AVG_SAMPLES_MAX)
-		return -EINVAL;
-
-	return __ffs(value);
-}
-
-static int adc5_decimation_from_dt(u32 value,
-					const unsigned int *decimation)
-{
-	unsigned int i;
-
-	for (i = 0; i < ADC5_DECIMATION_SAMPLES_MAX; i++) {
-		if (value == decimation[i])
-			return i;
-	}
-
-	return -EINVAL;
-}
-
 static int adc5_read_voltage_data(struct adc5_chip *adc, u16 *data)
 {
 	int ret;
@@ -511,7 +450,7 @@ static int adc_read_raw_common(struct iio_dev *indio_dev,
 			return ret;
 
 		ret = qcom_adc5_hw_scale(prop->scale_fn_type,
-			&adc5_prescale_ratios[prop->prescale],
+			prop->prescale,
 			adc->data,
 			adc_code_volt, val);
 		if (ret)
@@ -717,7 +656,7 @@ static int adc5_get_dt_channel_data(struct adc5_chip *adc,
 
 	ret = of_property_read_u32(node, "qcom,decimation", &value);
 	if (!ret) {
-		ret = adc5_decimation_from_dt(value, data->decimation);
+		ret = qcom_adc5_decimation_from_dt(value, data->decimation);
 		if (ret < 0) {
 			dev_err(dev, "%02x invalid decimation %d\n",
 				chan, value);
@@ -730,7 +669,7 @@ static int adc5_get_dt_channel_data(struct adc5_chip *adc,
 
 	ret = of_property_read_u32_array(node, "qcom,pre-scaling", varr, 2);
 	if (!ret) {
-		ret = adc5_prescaling_from_dt(varr[0], varr[1]);
+		ret = qcom_adc5_prescaling_from_dt(varr[0], varr[1]);
 		if (ret < 0) {
 			dev_err(dev, "%02x invalid pre-scaling <%d %d>\n",
 				chan, varr[0], varr[1]);
@@ -759,11 +698,9 @@ static int adc5_get_dt_channel_data(struct adc5_chip *adc,
 		if ((dig_version[0] >= ADC5_HW_SETTLE_DIFF_MINOR &&
 			dig_version[1] >= ADC5_HW_SETTLE_DIFF_MAJOR) ||
 			adc->data->info == &adc7_info)
-			ret = adc5_hw_settle_time_from_dt(value,
-							data->hw_settle_2);
+			ret = qcom_adc5_hw_settle_time_from_dt(value, data->hw_settle_2);
 		else
-			ret = adc5_hw_settle_time_from_dt(value,
-							data->hw_settle_1);
+			ret = qcom_adc5_hw_settle_time_from_dt(value, data->hw_settle_1);
 
 		if (ret < 0) {
 			dev_err(dev, "%02x invalid hw-settle-time %d us\n",
@@ -777,7 +714,7 @@ static int adc5_get_dt_channel_data(struct adc5_chip *adc,
 
 	ret = of_property_read_u32(node, "qcom,avg-samples", &value);
 	if (!ret) {
-		ret = adc5_avg_samples_from_dt(value);
+		ret = qcom_adc5_avg_samples_from_dt(value);
 		if (ret < 0) {
 			dev_err(dev, "%02x invalid avg-samples %d\n",
 				chan, value);
diff --git a/drivers/iio/adc/qcom-vadc-common.c b/drivers/iio/adc/qcom-vadc-common.c
index 5113aaa6ba67..d11f3343ad52 100644
--- a/drivers/iio/adc/qcom-vadc-common.c
+++ b/drivers/iio/adc/qcom-vadc-common.c
@@ -278,6 +278,18 @@ static const struct vadc_map_pt adcmap7_100k[] = {
 	{ 2420, 130048 }
 };
 
+static const struct vadc_prescale_ratio adc5_prescale_ratios[] = {
+	{.num =  1, .den =  1},
+	{.num =  1, .den =  3},
+	{.num =  1, .den =  4},
+	{.num =  1, .den =  6},
+	{.num =  1, .den = 20},
+	{.num =  1, .den =  8},
+	{.num = 10, .den = 81},
+	{.num =  1, .den = 10},
+	{.num =  1, .den = 16}
+};
+
 static int qcom_vadc_scale_hw_calib_volt(
 				const struct vadc_prescale_ratio *prescale,
 				const struct adc5_data *data,
@@ -647,10 +659,12 @@ int qcom_vadc_scale(enum vadc_scale_fn_type scaletype,
 EXPORT_SYMBOL(qcom_vadc_scale);
 
 int qcom_adc5_hw_scale(enum vadc_scale_fn_type scaletype,
-		    const struct vadc_prescale_ratio *prescale,
+		    unsigned int prescale_ratio,
 		    const struct adc5_data *data,
 		    u16 adc_code, int *result)
 {
+	const struct vadc_prescale_ratio *prescale = &adc5_prescale_ratios[prescale_ratio];
+
 	if (!(scaletype >= SCALE_HW_CALIB_DEFAULT &&
 		scaletype < SCALE_HW_CALIB_INVALID)) {
 		pr_err("Invalid scale type %d\n", scaletype);
@@ -662,6 +676,58 @@ int qcom_adc5_hw_scale(enum vadc_scale_fn_type scaletype,
 }
 EXPORT_SYMBOL(qcom_adc5_hw_scale);
 
+int qcom_adc5_prescaling_from_dt(u32 num, u32 den)
+{
+	unsigned int pre;
+
+	for (pre = 0; pre < ARRAY_SIZE(adc5_prescale_ratios); pre++)
+		if (adc5_prescale_ratios[pre].num == num &&
+		    adc5_prescale_ratios[pre].den == den)
+			break;
+
+	if (pre == ARRAY_SIZE(adc5_prescale_ratios))
+		return -EINVAL;
+
+	return pre;
+}
+EXPORT_SYMBOL(qcom_adc5_prescaling_from_dt);
+
+int qcom_adc5_hw_settle_time_from_dt(u32 value,
+				     const unsigned int *hw_settle)
+{
+	unsigned int i;
+
+	for (i = 0; i < VADC_HW_SETTLE_SAMPLES_MAX; i++) {
+		if (value == hw_settle[i])
+			return i;
+	}
+
+	return -EINVAL;
+}
+EXPORT_SYMBOL(qcom_adc5_hw_settle_time_from_dt);
+
+int qcom_adc5_avg_samples_from_dt(u32 value)
+{
+	if (!is_power_of_2(value) || value > ADC5_AVG_SAMPLES_MAX)
+		return -EINVAL;
+
+	return __ffs(value);
+}
+EXPORT_SYMBOL(qcom_adc5_avg_samples_from_dt);
+
+int qcom_adc5_decimation_from_dt(u32 value, const unsigned int *decimation)
+{
+	unsigned int i;
+
+	for (i = 0; i < ADC5_DECIMATION_SAMPLES_MAX; i++) {
+		if (value == decimation[i])
+			return i;
+	}
+
+	return -EINVAL;
+}
+EXPORT_SYMBOL(qcom_adc5_decimation_from_dt);
+
 int qcom_vadc_decimation_from_dt(u32 value)
 {
 	if (!is_power_of_2(value) || value < VADC_DECIMATION_MIN ||
diff --git a/drivers/iio/adc/qcom-vadc-common.h b/drivers/iio/adc/qcom-vadc-common.h
index 17b2fc4d8bf2..7e5f6428e311 100644
--- a/drivers/iio/adc/qcom-vadc-common.h
+++ b/drivers/iio/adc/qcom-vadc-common.h
@@ -168,10 +168,18 @@ struct qcom_adc5_scale_type {
 };
 
 int qcom_adc5_hw_scale(enum vadc_scale_fn_type scaletype,
-		    const struct vadc_prescale_ratio *prescale,
+		    unsigned int prescale_ratio,
 		    const struct adc5_data *data,
 		    u16 adc_code, int *result_mdec);
 
+int qcom_adc5_prescaling_from_dt(u32 num, u32 den);
+
+int qcom_adc5_hw_settle_time_from_dt(u32 value, const unsigned int *hw_settle);
+
+int qcom_adc5_avg_samples_from_dt(u32 value);
+
+int qcom_adc5_decimation_from_dt(u32 value, const unsigned int *decimation);
+
 int qcom_vadc_decimation_from_dt(u32 value);
 
 #endif /* QCOM_VADC_COMMON_H */
-- 
2.28.0


^ permalink raw reply related	[flat|nested] 15+ messages in thread

* [PATCH v5 4/9] iio: adc: qcom-vadc-common: use fixp_linear_interpolate
  2020-09-14 15:48 [PATCH v5 0/9] qcom: pm8150: add support for thermal monitoring Dmitry Baryshkov
                   ` (2 preceding siblings ...)
  2020-09-14 15:48 ` [PATCH v5 3/9] iio: adc: qcom-vadc: move several adc5 functions to common file Dmitry Baryshkov
@ 2020-09-14 15:48 ` Dmitry Baryshkov
  2020-09-14 15:48 ` [PATCH v5 5/9] iio: adc: qcom-spmi-adc5: use of_device_get_match_data Dmitry Baryshkov
                   ` (4 subsequent siblings)
  8 siblings, 0 replies; 15+ messages in thread
From: Dmitry Baryshkov @ 2020-09-14 15:48 UTC (permalink / raw)
  To: Andy Gross, Bjorn Andersson, Rob Herring, Zhang Rui,
	Daniel Lezcano, Amit Kucheria, Jonathan Cameron, Hartmut Knaack,
	Lars-Peter Clausen, Peter Meerwald-Stadler
  Cc: linux-arm-msm, linux-pm, devicetree, linux-iio,
	Manivannan Sadhasivam, Jonathan Cameron

Use new function fixp_linear_interpolate() instead of hand-coding the
linear interpolation.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
---
 drivers/iio/adc/qcom-vadc-common.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/iio/adc/qcom-vadc-common.c b/drivers/iio/adc/qcom-vadc-common.c
index d11f3343ad52..40d77b3af1bb 100644
--- a/drivers/iio/adc/qcom-vadc-common.c
+++ b/drivers/iio/adc/qcom-vadc-common.c
@@ -2,6 +2,7 @@
 #include <linux/bug.h>
 #include <linux/kernel.h>
 #include <linux/bitops.h>
+#include <linux/fixp-arith.h>
 #include <linux/math64.h>
 #include <linux/log2.h>
 #include <linux/err.h>
@@ -368,10 +369,9 @@ static int qcom_vadc_map_voltage_temp(const struct vadc_map_pt *pts,
 	} else {
 		/* result is between search_index and search_index-1 */
 		/* interpolate linearly */
-		*output = (((s32)((pts[i].y - pts[i - 1].y) *
-			(input - pts[i - 1].x)) /
-			(pts[i].x - pts[i - 1].x)) +
-			pts[i - 1].y);
+		*output = fixp_linear_interpolate(pts[i - 1].x, pts[i - 1].y,
+						  pts[i].x, pts[i].y,
+						  input);
 	}
 
 	return 0;
-- 
2.28.0


^ permalink raw reply related	[flat|nested] 15+ messages in thread

* [PATCH v5 5/9] iio: adc: qcom-spmi-adc5: use of_device_get_match_data
  2020-09-14 15:48 [PATCH v5 0/9] qcom: pm8150: add support for thermal monitoring Dmitry Baryshkov
                   ` (3 preceding siblings ...)
  2020-09-14 15:48 ` [PATCH v5 4/9] iio: adc: qcom-vadc-common: use fixp_linear_interpolate Dmitry Baryshkov
@ 2020-09-14 15:48 ` Dmitry Baryshkov
  2020-09-14 15:48 ` [PATCH v5 6/9] iio: provide of_iio_channel_get_by_name() and devm_ version it Dmitry Baryshkov
                   ` (3 subsequent siblings)
  8 siblings, 0 replies; 15+ messages in thread
From: Dmitry Baryshkov @ 2020-09-14 15:48 UTC (permalink / raw)
  To: Andy Gross, Bjorn Andersson, Rob Herring, Zhang Rui,
	Daniel Lezcano, Amit Kucheria, Jonathan Cameron, Hartmut Knaack,
	Lars-Peter Clausen, Peter Meerwald-Stadler
  Cc: linux-arm-msm, linux-pm, devicetree, linux-iio,
	Manivannan Sadhasivam, Jonathan Cameron

Use of_device_get_match_data() instead of hand-coding it manually.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Acked-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Acked-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
---
 drivers/iio/adc/qcom-spmi-adc5.c | 18 +++++++-----------
 1 file changed, 7 insertions(+), 11 deletions(-)

diff --git a/drivers/iio/adc/qcom-spmi-adc5.c b/drivers/iio/adc/qcom-spmi-adc5.c
index c2da8f068b87..ae27ed7a26ff 100644
--- a/drivers/iio/adc/qcom-spmi-adc5.c
+++ b/drivers/iio/adc/qcom-spmi-adc5.c
@@ -14,6 +14,7 @@
 #include <linux/math64.h>
 #include <linux/module.h>
 #include <linux/of.h>
+#include <linux/of_device.h>
 #include <linux/platform_device.h>
 #include <linux/regmap.h>
 #include <linux/slab.h>
@@ -807,8 +808,6 @@ static int adc5_get_dt_data(struct adc5_chip *adc, struct device_node *node)
 	struct adc5_channel_prop prop, *chan_props;
 	struct device_node *child;
 	unsigned int index = 0;
-	const struct of_device_id *id;
-	const struct adc5_data *data;
 	int ret;
 
 	adc->nchannels = of_get_available_child_count(node);
@@ -827,24 +826,21 @@ static int adc5_get_dt_data(struct adc5_chip *adc, struct device_node *node)
 
 	chan_props = adc->chan_props;
 	iio_chan = adc->iio_chans;
-	id = of_match_node(adc5_match_table, node);
-	if (id)
-		data = id->data;
-	else
-		data = &adc5_data_pmic;
-	adc->data = data;
+	adc->data = of_device_get_match_data(adc->dev);
+	if (!adc->data)
+		adc->data = &adc5_data_pmic;
 
 	for_each_available_child_of_node(node, child) {
-		ret = adc5_get_dt_channel_data(adc, &prop, child, data);
+		ret = adc5_get_dt_channel_data(adc, &prop, child, adc->data);
 		if (ret) {
 			of_node_put(child);
 			return ret;
 		}
 
 		prop.scale_fn_type =
-			data->adc_chans[prop.channel].scale_fn_type;
+			adc->data->adc_chans[prop.channel].scale_fn_type;
 		*chan_props = prop;
-		adc_chan = &data->adc_chans[prop.channel];
+		adc_chan = &adc->data->adc_chans[prop.channel];
 
 		iio_chan->channel = prop.channel;
 		iio_chan->datasheet_name = prop.datasheet_name;
-- 
2.28.0


^ permalink raw reply related	[flat|nested] 15+ messages in thread

* [PATCH v5 6/9] iio: provide of_iio_channel_get_by_name() and devm_ version it
  2020-09-14 15:48 [PATCH v5 0/9] qcom: pm8150: add support for thermal monitoring Dmitry Baryshkov
                   ` (4 preceding siblings ...)
  2020-09-14 15:48 ` [PATCH v5 5/9] iio: adc: qcom-spmi-adc5: use of_device_get_match_data Dmitry Baryshkov
@ 2020-09-14 15:48 ` Dmitry Baryshkov
  2020-09-14 15:48 ` [PATCH v5 7/9] thermal: qcom: add support for adc-tm5 PMIC thermal monitor Dmitry Baryshkov
                   ` (2 subsequent siblings)
  8 siblings, 0 replies; 15+ messages in thread
From: Dmitry Baryshkov @ 2020-09-14 15:48 UTC (permalink / raw)
  To: Andy Gross, Bjorn Andersson, Rob Herring, Zhang Rui,
	Daniel Lezcano, Amit Kucheria, Jonathan Cameron, Hartmut Knaack,
	Lars-Peter Clausen, Peter Meerwald-Stadler
  Cc: linux-arm-msm, linux-pm, devicetree, linux-iio,
	Manivannan Sadhasivam, Jonathan Cameron

There might be cases when the IIO channel is attached to the device
subnode instead of being attached to the main device node. Allow drivers
to query IIO channels by using device tree nodes.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
---
 drivers/iio/inkern.c         | 34 ++++++++++++++++++++++++++--------
 include/linux/iio/consumer.h | 36 ++++++++++++++++++++++++++++++++++++
 2 files changed, 62 insertions(+), 8 deletions(-)

diff --git a/drivers/iio/inkern.c b/drivers/iio/inkern.c
index ede99e0d5371..8ff0ac472de3 100644
--- a/drivers/iio/inkern.c
+++ b/drivers/iio/inkern.c
@@ -180,8 +180,8 @@ static struct iio_channel *of_iio_channel_get(struct device_node *np, int index)
 	return ERR_PTR(err);
 }
 
-static struct iio_channel *of_iio_channel_get_by_name(struct device_node *np,
-						      const char *name)
+struct iio_channel *of_iio_channel_get_by_name(struct device_node *np,
+					       const char *name)
 {
 	struct iio_channel *chan = NULL;
 
@@ -219,6 +219,7 @@ static struct iio_channel *of_iio_channel_get_by_name(struct device_node *np,
 
 	return chan;
 }
+EXPORT_SYMBOL_GPL(of_iio_channel_get_by_name);
 
 static struct iio_channel *of_iio_channel_get_all(struct device *dev)
 {
@@ -261,12 +262,6 @@ static struct iio_channel *of_iio_channel_get_all(struct device *dev)
 
 #else /* CONFIG_OF */
 
-static inline struct iio_channel *
-of_iio_channel_get_by_name(struct device_node *np, const char *name)
-{
-	return NULL;
-}
-
 static inline struct iio_channel *of_iio_channel_get_all(struct device *dev)
 {
 	return NULL;
@@ -382,6 +377,29 @@ struct iio_channel *devm_iio_channel_get(struct device *dev,
 }
 EXPORT_SYMBOL_GPL(devm_iio_channel_get);
 
+struct iio_channel *devm_of_iio_channel_get_by_name(struct device *dev,
+						    struct device_node *np,
+						    const char *channel_name)
+{
+	struct iio_channel **ptr, *channel;
+
+	ptr = devres_alloc(devm_iio_channel_free, sizeof(*ptr), GFP_KERNEL);
+	if (!ptr)
+		return ERR_PTR(-ENOMEM);
+
+	channel = of_iio_channel_get_by_name(np, channel_name);
+	if (IS_ERR(channel)) {
+		devres_free(ptr);
+		return channel;
+	}
+
+	*ptr = channel;
+	devres_add(dev, ptr);
+
+	return channel;
+}
+EXPORT_SYMBOL_GPL(devm_of_iio_channel_get_by_name);
+
 struct iio_channel *iio_channel_get_all(struct device *dev)
 {
 	const char *name;
diff --git a/include/linux/iio/consumer.h b/include/linux/iio/consumer.h
index c4118dcb8e05..0a90ba8fa1bb 100644
--- a/include/linux/iio/consumer.h
+++ b/include/linux/iio/consumer.h
@@ -13,6 +13,7 @@
 struct iio_dev;
 struct iio_chan_spec;
 struct device;
+struct device_node;
 
 /**
  * struct iio_channel - everything needed for a consumer to use a channel
@@ -97,6 +98,41 @@ void iio_channel_release_all(struct iio_channel *chan);
  */
 struct iio_channel *devm_iio_channel_get_all(struct device *dev);
 
+/**
+ * of_iio_channel_get_by_name() - get description of all that is needed to access channel.
+ * @np:			Pointer to consumer device tree node
+ * @consumer_channel:	Unique name to identify the channel on the consumer
+ *			side. This typically describes the channels use within
+ *			the consumer. E.g. 'battery_voltage'
+ */
+#ifdef CONFIG_OF
+struct iio_channel *of_iio_channel_get_by_name(struct device_node *np, const char *name);
+#else
+static inline struct iio_channel *
+of_iio_channel_get_by_name(struct device_node *np, const char *name)
+{
+	return NULL;
+}
+#endif
+
+/**
+ * devm_of_iio_channel_get_by_name() - Resource managed version of of_iio_channel_get_by_name().
+ * @dev:		Pointer to consumer device.
+ * @np:			Pointer to consumer device tree node
+ * @consumer_channel:	Unique name to identify the channel on the consumer
+ *			side. This typically describes the channels use within
+ *			the consumer. E.g. 'battery_voltage'
+ *
+ * Returns a pointer to negative errno if it is not able to get the iio channel
+ * otherwise returns valid pointer for iio channel.
+ *
+ * The allocated iio channel is automatically released when the device is
+ * unbound.
+ */
+struct iio_channel *devm_of_iio_channel_get_by_name(struct device *dev,
+						    struct device_node *np,
+						    const char *consumer_channel);
+
 struct iio_cb_buffer;
 /**
  * iio_channel_get_all_cb() - register callback for triggered capture
-- 
2.28.0


^ permalink raw reply related	[flat|nested] 15+ messages in thread

* [PATCH v5 7/9] thermal: qcom: add support for adc-tm5 PMIC thermal monitor
  2020-09-14 15:48 [PATCH v5 0/9] qcom: pm8150: add support for thermal monitoring Dmitry Baryshkov
                   ` (5 preceding siblings ...)
  2020-09-14 15:48 ` [PATCH v5 6/9] iio: provide of_iio_channel_get_by_name() and devm_ version it Dmitry Baryshkov
@ 2020-09-14 15:48 ` Dmitry Baryshkov
  2020-09-25 11:42   ` Jishnu Prakash
  2020-09-14 15:48 ` [PATCH v5 8/9] arm64: dts: qcom: pm8150x: add definitions for adc-tm5 part Dmitry Baryshkov
  2020-09-14 15:48 ` [PATCH v5 9/9] arm64: dts: sm8250-mtp: add thermal zones using pmic's adc-tm5 Dmitry Baryshkov
  8 siblings, 1 reply; 15+ messages in thread
From: Dmitry Baryshkov @ 2020-09-14 15:48 UTC (permalink / raw)
  To: Andy Gross, Bjorn Andersson, Rob Herring, Zhang Rui,
	Daniel Lezcano, Amit Kucheria, Jonathan Cameron, Hartmut Knaack,
	Lars-Peter Clausen, Peter Meerwald-Stadler
  Cc: linux-arm-msm, linux-pm, devicetree, linux-iio,
	Manivannan Sadhasivam, Jonathan Cameron

Add support for Thermal Monitoring part of PMIC5. This part is closely
coupled with ADC, using it's channels directly. ADC-TM support
generating interrupts on ADC value crossing low or high voltage bounds,
which is used to support thermal trip points.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
---
 drivers/iio/adc/qcom-vadc-common.c       |  62 +++
 drivers/iio/adc/qcom-vadc-common.h       |   3 +
 drivers/thermal/qcom/Kconfig             |  11 +
 drivers/thermal/qcom/Makefile            |   1 +
 drivers/thermal/qcom/qcom-spmi-adc-tm5.c | 583 +++++++++++++++++++++++
 5 files changed, 660 insertions(+)
 create mode 100644 drivers/thermal/qcom/qcom-spmi-adc-tm5.c

diff --git a/drivers/iio/adc/qcom-vadc-common.c b/drivers/iio/adc/qcom-vadc-common.c
index 40d77b3af1bb..e58e393b8713 100644
--- a/drivers/iio/adc/qcom-vadc-common.c
+++ b/drivers/iio/adc/qcom-vadc-common.c
@@ -377,6 +377,42 @@ static int qcom_vadc_map_voltage_temp(const struct vadc_map_pt *pts,
 	return 0;
 }
 
+static s32 qcom_vadc_map_temp_voltage(const struct vadc_map_pt *pts,
+				      u32 tablesize, int input)
+{
+	bool descending = 1;
+	u32 i = 0;
+
+	/* Check if table is descending or ascending */
+	if (tablesize > 1) {
+		if (pts[0].y < pts[1].y)
+			descending = 0;
+	}
+
+	while (i < tablesize) {
+		if (descending && pts[i].y < input) {
+			/* table entry is less than measured*/
+			 /* value and table is descending, stop */
+			break;
+		} else if ((!descending) && pts[i].y > input) {
+			/* table entry is greater than measured*/
+			/*value and table is ascending, stop */
+			break;
+		}
+		i++;
+	}
+
+	if (i == 0)
+		return pts[0].x;
+	if (i == tablesize)
+		return pts[tablesize - 1].x;
+
+	/* result is between search_index and search_index-1 */
+	/* interpolate linearly */
+	return fixp_linear_interpolate(pts[i - 1].y, pts[i - 1].x,
+			pts[i].y, pts[i].x, input);
+}
+
 static void qcom_vadc_scale_calib(const struct vadc_linear_graph *calib_graph,
 				  u16 adc_code,
 				  bool absolute,
@@ -474,6 +510,19 @@ static int qcom_vadc_scale_chg_temp(const struct vadc_linear_graph *calib_graph,
 	return 0;
 }
 
+static u16 qcom_vadc_scale_voltage_code(int voltage,
+					const struct vadc_prescale_ratio *prescale,
+					const u32 full_scale_code_volt,
+					unsigned int factor)
+{
+	s64 volt = voltage, adc_vdd_ref_mv = 1875;
+
+	volt *= prescale->num * factor * full_scale_code_volt;
+	volt = div64_s64(volt, (s64)prescale->den * adc_vdd_ref_mv * 1000);
+
+	return volt;
+}
+
 static int qcom_vadc_scale_code_voltage_factor(u16 adc_code,
 				const struct vadc_prescale_ratio *prescale,
 				const struct adc5_data *data,
@@ -658,6 +707,19 @@ int qcom_vadc_scale(enum vadc_scale_fn_type scaletype,
 }
 EXPORT_SYMBOL(qcom_vadc_scale);
 
+u16 qcom_adc_tm5_temp_volt_scale(unsigned int prescale_ratio,
+				 u32 full_scale_code_volt, int temp)
+{
+	const struct vadc_prescale_ratio *prescale = &adc5_prescale_ratios[prescale_ratio];
+	s32 voltage;
+
+	voltage = qcom_vadc_map_temp_voltage(adcmap_100k_104ef_104fb_1875_vref,
+					     ARRAY_SIZE(adcmap_100k_104ef_104fb_1875_vref),
+					     temp);
+	return qcom_vadc_scale_voltage_code(voltage, prescale, full_scale_code_volt, 1000);
+}
+EXPORT_SYMBOL(qcom_adc_tm5_temp_volt_scale);
+
 int qcom_adc5_hw_scale(enum vadc_scale_fn_type scaletype,
 		    unsigned int prescale_ratio,
 		    const struct adc5_data *data,
diff --git a/drivers/iio/adc/qcom-vadc-common.h b/drivers/iio/adc/qcom-vadc-common.h
index 7e5f6428e311..9af41201ad77 100644
--- a/drivers/iio/adc/qcom-vadc-common.h
+++ b/drivers/iio/adc/qcom-vadc-common.h
@@ -172,6 +172,9 @@ int qcom_adc5_hw_scale(enum vadc_scale_fn_type scaletype,
 		    const struct adc5_data *data,
 		    u16 adc_code, int *result_mdec);
 
+u16 qcom_adc_tm5_temp_volt_scale(unsigned int prescale_ratio,
+				 u32 full_scale_code_volt, int temp);
+
 int qcom_adc5_prescaling_from_dt(u32 num, u32 den);
 
 int qcom_adc5_hw_settle_time_from_dt(u32 value, const unsigned int *hw_settle);
diff --git a/drivers/thermal/qcom/Kconfig b/drivers/thermal/qcom/Kconfig
index aa9c1d80fae4..8d5ac2df26dc 100644
--- a/drivers/thermal/qcom/Kconfig
+++ b/drivers/thermal/qcom/Kconfig
@@ -10,6 +10,17 @@ config QCOM_TSENS
 	  Also able to set threshold temperature for both hot and cold and update
 	  when a threshold is reached.
 
+config QCOM_SPMI_ADC_TM5
+	tristate "Qualcomm SPMI PMIC Thermal Monitor ADC5"
+	depends on OF && SPMI && IIO
+	select REGMAP_SPMI
+	select QCOM_VADC_COMMON
+	help
+	  This enables the thermal driver for the ADC thermal monitoring
+	  device. It shows up as a thermal zone with multiple trip points.
+	  Thermal client sets threshold temperature for both warm and cool and
+	  gets updated when a threshold is reached.
+
 config QCOM_SPMI_TEMP_ALARM
 	tristate "Qualcomm SPMI PMIC Temperature Alarm"
 	depends on OF && SPMI && IIO
diff --git a/drivers/thermal/qcom/Makefile b/drivers/thermal/qcom/Makefile
index ec86eef7f6a6..252ea7d9da0b 100644
--- a/drivers/thermal/qcom/Makefile
+++ b/drivers/thermal/qcom/Makefile
@@ -3,4 +3,5 @@ obj-$(CONFIG_QCOM_TSENS)	+= qcom_tsens.o
 
 qcom_tsens-y			+= tsens.o tsens-v2.o tsens-v1.o tsens-v0_1.o \
 				   tsens-8960.o
+obj-$(CONFIG_QCOM_SPMI_ADC_TM5)	+= qcom-spmi-adc-tm5.o
 obj-$(CONFIG_QCOM_SPMI_TEMP_ALARM)	+= qcom-spmi-temp-alarm.o
diff --git a/drivers/thermal/qcom/qcom-spmi-adc-tm5.c b/drivers/thermal/qcom/qcom-spmi-adc-tm5.c
new file mode 100644
index 000000000000..bfb403d20226
--- /dev/null
+++ b/drivers/thermal/qcom/qcom-spmi-adc-tm5.c
@@ -0,0 +1,583 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2012-2020, The Linux Foundation. All rights reserved.
+ * Copyright (c) 2020 Linaro Limited
+ */
+
+#include <linux/iio/consumer.h>
+#include <linux/interrupt.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
+#include <linux/regmap.h>
+#include <linux/thermal.h>
+
+#include "../../iio/adc/qcom-vadc-common.h"
+
+#define ADC5_MAX_CHANNEL                        0xc0
+#define ADC_TM5_NUM_CHANNELS		8
+
+#define ADC_TM5_STATUS_LOW			0x0a
+
+#define ADC_TM5_STATUS_HIGH			0x0b
+
+#define ADC_TM5_NUM_BTM				0x0f
+
+#define ADC_TM5_ADC_DIG_PARAM			0x42
+
+#define ADC_TM5_FAST_AVG_CTL			(ADC_TM5_ADC_DIG_PARAM + 1)
+#define ADC_TM5_FAST_AVG_EN				BIT(7)
+
+#define ADC_TM5_MEAS_INTERVAL_CTL		(ADC_TM5_ADC_DIG_PARAM + 2)
+#define ADC_TM5_TIMER1					3 /* 3.9ms */
+
+#define ADC_TM5_MEAS_INTERVAL_CTL2		(ADC_TM5_ADC_DIG_PARAM + 3)
+#define ADC_TM5_MEAS_INTERVAL_CTL2_MASK			0xf0
+#define ADC_TM5_TIMER2					10 /* 1 second */
+#define ADC_TM5_MEAS_INTERVAL_CTL3_MASK			0xf
+#define ADC_TM5_TIMER3					4 /* 4 second */
+
+#define ADC_TM5_M_CHAN_BASE			0x60
+
+#define ADC_TM5_M_ADC_CH_SEL_CTL(n)		(ADC_TM5_M_CHAN_BASE + ((n) * 8) + 0)
+#define ADC_TM5_M_LOW_THR0(n)			(ADC_TM5_M_CHAN_BASE + ((n) * 8) + 1)
+#define ADC_TM5_M_LOW_THR1(n)			(ADC_TM5_M_CHAN_BASE + ((n) * 8) + 2)
+#define ADC_TM5_M_HIGH_THR0(n)			(ADC_TM5_M_CHAN_BASE + ((n) * 8) + 3)
+#define ADC_TM5_M_HIGH_THR1(n)			(ADC_TM5_M_CHAN_BASE + ((n) * 8) + 4)
+#define ADC_TM5_M_MEAS_INTERVAL_CTL(n)		(ADC_TM5_M_CHAN_BASE + ((n) * 8) + 5)
+#define ADC_TM5_M_CTL(n)			(ADC_TM5_M_CHAN_BASE + ((n) * 8) + 6)
+#define ADC_TM5_M_CTL_HW_SETTLE_DELAY_MASK		0xf
+#define ADC_TM5_M_CTL_CAL_SEL_MASK			0x30
+#define ADC_TM5_M_CTL_CAL_VAL				0x40
+#define ADC_TM5_M_EN(n)				(ADC_TM5_M_CHAN_BASE + ((n) * 8) + 7)
+#define ADC_TM5_M_MEAS_EN				BIT(7)
+#define ADC_TM5_M_HIGH_THR_INT_EN			BIT(1)
+#define ADC_TM5_M_LOW_THR_INT_EN			BIT(0)
+
+enum adc5_timer_select {
+	ADC5_TIMER_SEL_1 = 0,
+	ADC5_TIMER_SEL_2,
+	ADC5_TIMER_SEL_3,
+	ADC5_TIMER_SEL_NONE,
+};
+
+struct adc_tm5_data {
+	const u32	full_scale_code_volt;
+	unsigned int	*decimation;
+	unsigned int	*hw_settle;
+};
+
+enum adc_tm5_cal_method {
+	ADC_TM5_NO_CAL = 0,
+	ADC_TM5_RATIOMETRIC_CAL,
+	ADC_TM5_ABSOLUTE_CAL
+};
+
+struct adc_tm5_chip;
+
+/**
+ * struct adc_tm5_channel - ADC Thermal Monitoring channel data.
+ * @channel: channel number.
+ * @adc_channel: corresponding ADC channel number.
+ * @cal_method: calibration method.
+ * @prescale: channel scaling performed on the input signal.
+ * @hw_settle_time: the time between AMUX being configured and the
+ *	start of conversion.
+ * @iio: IIO channel instance used by this channel.
+ * @chip: ADC TM chip instance.
+ * @tzd: thermal zone device used by this channel.
+ */
+struct adc_tm5_channel {
+	unsigned int		channel;
+	unsigned int		adc_channel;
+	enum adc_tm5_cal_method	cal_method;
+	unsigned int		prescale;
+	unsigned int		hw_settle_time;
+	struct iio_channel	*iio;
+	struct adc_tm5_chip	*chip;
+	struct thermal_zone_device *tzd;
+};
+
+/**
+ * struct adc_tm5_chip - ADC Thermal Monitoring properties
+ * @regmap: SPMI ADC5 Thermal Monitoring  peripheral register map field.
+ * @dev: SPMI ADC5 device.
+ * @data: software configuration data.
+ * @channels: array of ADC TM channel data.
+ * @nchannels: amount of channels defined/allocated
+ * @decimation: sampling rate supported for the channel.
+ * @avg_samples: ability to provide single result from the ADC
+ *	that is an average of multiple measurements.
+ * @base: base address of TM registers.
+ */
+struct adc_tm5_chip {
+	struct regmap		*regmap;
+	struct device		*dev;
+	const struct adc_tm5_data	*data;
+	struct adc_tm5_channel	*channels;
+	unsigned int		nchannels;
+	unsigned int		decimation;
+	unsigned int		avg_samples;
+	u16			base;
+};
+
+static const struct adc_tm5_data adc_tm5_data_pmic = {
+	.full_scale_code_volt = 0x70e4,
+	.decimation = (unsigned int []) {250, 420, 840},
+	.hw_settle = (unsigned int []) {15, 100, 200, 300, 400, 500, 600, 700,
+					1, 2, 4, 8, 16, 32, 64, 128},
+};
+
+static int adc_tm5_read(struct adc_tm5_chip *adc_tm, u16 offset, u8 *data, int len)
+{
+	return regmap_bulk_read(adc_tm->regmap, adc_tm->base + offset, data, len);
+}
+
+static int adc_tm5_write(struct adc_tm5_chip *adc_tm, u16 offset, u8 *data, int len)
+{
+	return regmap_bulk_write(adc_tm->regmap, adc_tm->base + offset, data, len);
+}
+
+static int adc_tm5_reg_update(struct adc_tm5_chip *adc_tm, u16 offset, u8 mask, u8 val)
+{
+	return regmap_write_bits(adc_tm->regmap, adc_tm->base + offset, mask, val);
+}
+
+static irqreturn_t adc_tm5_isr(int irq, void *data)
+{
+	struct adc_tm5_chip *chip = data;
+	u8 status_low, status_high, ctl;
+	int ret = 0, i = 0;
+
+	ret = adc_tm5_read(chip, ADC_TM5_STATUS_LOW, &status_low, 1);
+	if (ret) {
+		dev_err(chip->dev, "read status low failed with %d\n", ret);
+		return IRQ_HANDLED;
+	}
+
+	ret = adc_tm5_read(chip, ADC_TM5_STATUS_HIGH, &status_high, 1);
+	if (ret) {
+		dev_err(chip->dev, "read status high failed with %d\n", ret);
+		return IRQ_HANDLED;
+	}
+
+	for (i = 0; i < chip->nchannels; i++) {
+		bool upper_set = false, lower_set = false;
+		unsigned int ch = chip->channels[i].channel;
+
+		if (!chip->channels[i].tzd) {
+			dev_err_once(chip->dev, "thermal device not found\n");
+			continue;
+		}
+
+		ret = adc_tm5_read(chip, ADC_TM5_M_EN(ch), &ctl, 1);
+
+		if (ret) {
+			dev_err(chip->dev, "ctl read failed with %d\n", ret);
+			continue;
+		}
+
+		lower_set = (status_low & BIT(ch)) &&
+			(ctl & ADC_TM5_M_MEAS_EN) &&
+			(ctl & ADC_TM5_M_LOW_THR_INT_EN);
+
+		upper_set = (status_high & BIT(ch)) &&
+			(ctl & ADC_TM5_M_MEAS_EN) &&
+			(ctl & ADC_TM5_M_HIGH_THR_INT_EN);
+
+		if (upper_set || lower_set)
+			thermal_zone_device_update(chip->channels[i].tzd,
+						   THERMAL_EVENT_UNSPECIFIED);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static int adc_tm5_get_temp(void *data, int *temp)
+{
+	struct adc_tm5_channel *channel = data;
+	int ret, milli_celsius;
+
+	if (!channel || !channel->iio)
+		return -EINVAL;
+
+	ret = iio_read_channel_processed(channel->iio, &milli_celsius);
+	if (ret < 0)
+		return ret;
+
+	*temp = milli_celsius;
+
+	return 0;
+}
+
+static int adc_tm5_disable_channel(struct adc_tm5_channel *channel)
+{
+	struct adc_tm5_chip *chip = channel->chip;
+	unsigned int reg = ADC_TM5_M_EN(channel->channel);
+
+	return adc_tm5_reg_update(chip, reg,
+			ADC_TM5_M_MEAS_EN | ADC_TM5_M_HIGH_THR_INT_EN | ADC_TM5_M_LOW_THR_INT_EN,
+			0);
+}
+
+static int adc_tm5_configure(struct adc_tm5_channel *channel, int low_temp, int high_temp)
+{
+	struct adc_tm5_chip *chip = channel->chip;
+	u8 buf[8];
+	u16 reg = ADC_TM5_M_ADC_CH_SEL_CTL(channel->channel);
+	int ret = 0;
+
+	ret = adc_tm5_read(chip, reg, buf, sizeof(buf));
+	if (ret) {
+		dev_err(chip->dev, "block read failed with %d\n", ret);
+		return ret;
+	}
+
+	/* Update ADC channel select */
+	buf[0] = channel->adc_channel;
+
+	/* Warm temperature corresponds to low voltage threshold */
+	if (high_temp != INT_MAX) {
+		u16 adc_code = qcom_adc_tm5_temp_volt_scale(channel->prescale,
+				chip->data->full_scale_code_volt, high_temp);
+
+		buf[1] = adc_code & 0xff;
+		buf[2] = adc_code >> 8;
+		buf[7] |= ADC_TM5_M_LOW_THR_INT_EN;
+	}
+
+	/* Cool temperature corresponds to high voltage threshold */
+	if (low_temp != -INT_MAX) {
+		u16 adc_code = qcom_adc_tm5_temp_volt_scale(channel->prescale,
+				chip->data->full_scale_code_volt, low_temp);
+
+		buf[3] = adc_code & 0xff;
+		buf[4] = adc_code >> 8;
+		buf[7] |= ADC_TM5_M_HIGH_THR_INT_EN;
+	}
+
+	/* Update timer select */
+	buf[5] = ADC5_TIMER_SEL_2;
+
+	/* Set calibration select, hw_settle delay */
+	buf[6] &= ~ADC_TM5_M_CTL_HW_SETTLE_DELAY_MASK;
+	buf[6] |= FIELD_PREP(ADC_TM5_M_CTL_HW_SETTLE_DELAY_MASK, channel->hw_settle_time);
+	buf[6] &= ~ADC_TM5_M_CTL_CAL_SEL_MASK;
+	buf[6] |= FIELD_PREP(ADC_TM5_M_CTL_CAL_SEL_MASK, channel->cal_method);
+
+	buf[7] |= ADC_TM5_M_MEAS_EN;
+
+	ret = adc_tm5_write(chip, reg, buf, sizeof(buf));
+	if (ret)
+		dev_err(chip->dev, "buf write failed\n");
+
+	return ret;
+}
+
+static int adc_tm5_set_trips(void *data, int low_temp, int high_temp)
+{
+	struct adc_tm5_channel *channel = data;
+	struct adc_tm5_chip *chip;
+	int ret;
+
+	if (!channel)
+		return -EINVAL;
+
+	chip = channel->chip;
+	dev_dbg(chip->dev, "%d:low_temp(mdegC):%d, high_temp(mdegC):%d\n",
+		channel->channel, low_temp, high_temp);
+
+	if (high_temp == INT_MAX && low_temp <= -INT_MAX)
+		ret = adc_tm5_disable_channel(channel);
+	else
+		ret = adc_tm5_configure(channel, low_temp, high_temp);
+
+	return ret;
+}
+
+static struct thermal_zone_of_device_ops adc_tm5_ops = {
+	.get_temp = adc_tm5_get_temp,
+	.set_trips = adc_tm5_set_trips,
+};
+
+static int adc_tm5_register_tzd(struct adc_tm5_chip *adc_tm)
+{
+	unsigned int i;
+	struct thermal_zone_device *tzd;
+
+	for (i = 0; i < adc_tm->nchannels; i++) {
+		adc_tm->channels[i].chip = adc_tm;
+
+		tzd = devm_thermal_zone_of_sensor_register(adc_tm->dev,
+							   adc_tm->channels[i].channel,
+							   &adc_tm->channels[i],
+							   &adc_tm5_ops);
+		if (IS_ERR(tzd)) {
+			dev_err(adc_tm->dev, "Error registering TZ zone:%ld for channel:%d\n",
+				PTR_ERR(tzd), adc_tm->channels[i].channel);
+			continue;
+		}
+		adc_tm->channels[i].tzd = tzd;
+	}
+
+	return 0;
+}
+
+static int adc_tm5_init(struct adc_tm5_chip *chip)
+{
+	u8 buf[4], channels_available;
+	int ret;
+	unsigned int i;
+
+	ret = adc_tm5_read(chip, ADC_TM5_NUM_BTM, &channels_available, 1);
+	if (ret) {
+		dev_err(chip->dev, "read failed for BTM channels\n");
+		return ret;
+	}
+
+	ret = adc_tm5_read(chip, ADC_TM5_ADC_DIG_PARAM, buf, sizeof(buf));
+	if (ret) {
+		dev_err(chip->dev, "block read failed with %d\n", ret);
+		return ret;
+	}
+
+	/* Select decimation */
+	buf[0] = chip->decimation;
+
+	/* Select number of samples in fast average mode */
+	buf[1] = chip->avg_samples | ADC_TM5_FAST_AVG_EN;
+
+	/* Select timer1 */
+	buf[2] = ADC_TM5_TIMER1;
+
+	/* Select timer2 and timer3 */
+	buf[3] = FIELD_PREP(ADC_TM5_MEAS_INTERVAL_CTL2_MASK, ADC_TM5_TIMER2) |
+		 FIELD_PREP(ADC_TM5_MEAS_INTERVAL_CTL3_MASK, ADC_TM5_TIMER3);
+
+	ret = adc_tm5_write(chip, ADC_TM5_ADC_DIG_PARAM, buf, sizeof(buf));
+	if (ret)
+		dev_err(chip->dev, "block write failed with %d\n", ret);
+
+	for (i = 0; i < chip->nchannels; i++) {
+		if (chip->channels[i].channel >= channels_available) {
+			dev_err(chip->dev, "Invalid channel %d\n", chip->channels[i].channel);
+			return -EINVAL;
+		}
+	}
+
+	return ret;
+}
+
+static int adc_tm5_get_dt_channel_data(struct adc_tm5_chip *adc_tm,
+				       struct adc_tm5_channel *channel,
+				       struct device_node *node)
+{
+	const char *name = node->name;
+	u32 chan, value, varr[2];
+	int ret;
+	struct device *dev = adc_tm->dev;
+
+	ret = of_property_read_u32(node, "reg", &chan);
+	if (ret) {
+		dev_err(dev, "invalid channel number %s\n", name);
+		return ret;
+	}
+
+	if (chan >= ADC_TM5_NUM_CHANNELS) {
+		dev_err(dev, "%s invalid channel number %d\n", name, chan);
+		return -EINVAL;
+	}
+
+	/* the channel has DT description */
+	channel->channel = chan;
+
+	ret = of_property_read_u32(node, "qcom,adc-channel", &chan);
+	if (ret) {
+		dev_err(dev, "invalid channel number %s\n", name);
+		return ret;
+	}
+	if (chan >= ADC5_MAX_CHANNEL) {
+		dev_err(dev, "%s invalid ADC channel number %d\n", name, chan);
+		return ret;
+	}
+	channel->adc_channel = chan;
+
+	channel->iio = devm_of_iio_channel_get_by_name(adc_tm->dev, node, NULL);
+	if (IS_ERR(channel->iio)) {
+		ret = PTR_ERR(channel->iio);
+		channel->iio = NULL;
+		if (ret != -EPROBE_DEFER)
+			dev_err(dev, "error getting channel %s: %d\n", name, ret);
+		return ret;
+	}
+
+	ret = of_property_read_u32_array(node, "qcom,pre-scaling", varr, 2);
+	if (!ret) {
+		ret = qcom_adc5_prescaling_from_dt(varr[0], varr[1]);
+		if (ret < 0) {
+			dev_err(dev, "%02x invalid pre-scaling <%d %d>\n",
+				chan, varr[0], varr[1]);
+			return ret;
+		}
+		channel->prescale = ret;
+	} else {
+		/* 1:1 prescale is index 0 */
+		channel->prescale = 0;
+	}
+
+	ret = of_property_read_u32(node, "qcom,hw-settle-time", &value);
+	if (!ret) {
+		ret = qcom_adc5_hw_settle_time_from_dt(value, adc_tm->data->hw_settle);
+		if (ret < 0) {
+			dev_err(dev, "%02x invalid hw-settle-time %d us\n",
+				chan, value);
+			return ret;
+		}
+		channel->hw_settle_time = ret;
+	} else {
+		channel->hw_settle_time = VADC_DEF_HW_SETTLE_TIME;
+	}
+
+	if (of_property_read_bool(node, "qcom,ratiometric"))
+		channel->cal_method = ADC_TM5_RATIOMETRIC_CAL;
+	else
+		channel->cal_method = ADC_TM5_ABSOLUTE_CAL;
+
+	return 0;
+}
+
+static int adc_tm5_get_dt_data(struct adc_tm5_chip *adc_tm, struct device_node *node)
+{
+	struct adc_tm5_channel *channels;
+	struct device_node *child;
+	unsigned int index = 0;
+	u32 value;
+	int ret;
+	struct device *dev = adc_tm->dev;
+
+	adc_tm->nchannels = of_get_available_child_count(node);
+	if (!adc_tm->nchannels)
+		return -EINVAL;
+
+	adc_tm->channels = devm_kcalloc(dev, adc_tm->nchannels,
+					sizeof(*adc_tm->channels), GFP_KERNEL);
+	if (!adc_tm->channels)
+		return -ENOMEM;
+
+	channels = adc_tm->channels;
+
+	adc_tm->data = of_device_get_match_data(dev);
+	if (!adc_tm->data)
+		adc_tm->data = &adc_tm5_data_pmic;
+
+	ret = of_property_read_u32(node, "qcom,decimation", &value);
+	if (!ret) {
+		ret = qcom_adc5_decimation_from_dt(value, adc_tm->data->decimation);
+		if (ret < 0) {
+			dev_err(dev, "invalid decimation %d\n", value);
+			return ret;
+		}
+		adc_tm->decimation = ret;
+	} else {
+		adc_tm->decimation = ADC5_DECIMATION_DEFAULT;
+	}
+
+	ret = of_property_read_u32(node, "qcom,avg-samples", &value);
+	if (!ret) {
+		ret = qcom_adc5_avg_samples_from_dt(value);
+		if (ret < 0) {
+			dev_err(dev, "invalid avg-samples %d\n", value);
+			return ret;
+		}
+		adc_tm->avg_samples = ret;
+	} else {
+		adc_tm->avg_samples = VADC_DEF_AVG_SAMPLES;
+	}
+
+	for_each_available_child_of_node(node, child) {
+		ret = adc_tm5_get_dt_channel_data(adc_tm, channels, child);
+		if (ret) {
+			of_node_put(child);
+			return ret;
+		}
+
+		channels++;
+		index++;
+	}
+
+	return 0;
+}
+
+static int adc_tm5_probe(struct platform_device *pdev)
+{
+	struct device_node *node = pdev->dev.of_node;
+	struct device *dev = &pdev->dev;
+	struct adc_tm5_chip *adc_tm;
+	struct regmap *regmap;
+	int ret, irq;
+	u32 reg;
+
+	regmap = dev_get_regmap(dev->parent, NULL);
+	if (!regmap)
+		return -ENODEV;
+
+	ret = of_property_read_u32(node, "reg", &reg);
+	if (ret)
+		return ret;
+
+	adc_tm = devm_kzalloc(&pdev->dev, sizeof(*adc_tm), GFP_KERNEL);
+	if (!adc_tm)
+		return -ENOMEM;
+
+	adc_tm->regmap = regmap;
+	adc_tm->dev = dev;
+	adc_tm->base = reg;
+
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0) {
+		dev_err(dev, "get_irq failed: %d\n", irq);
+		return irq;
+	}
+
+	ret = adc_tm5_get_dt_data(adc_tm, node);
+	if (ret) {
+		dev_err(dev, "get dt data failed: %d\n", ret);
+		return ret;
+	}
+
+	ret = adc_tm5_init(adc_tm);
+	if (ret) {
+		dev_err(dev, "adc-tm init failed\n");
+		return ret;
+	}
+
+	ret = adc_tm5_register_tzd(adc_tm);
+	if (ret) {
+		dev_err(dev, "tzd register failed\n");
+		return ret;
+	}
+
+	return devm_request_irq(dev, irq, adc_tm5_isr, 0, "pm-adc-tm5", adc_tm);
+}
+
+static const struct of_device_id adc_tm5_match_table[] = {
+	{
+		.compatible = "qcom,spmi-adc-tm5",
+		.data = &adc_tm5_data_pmic,
+	},
+	{ }
+};
+MODULE_DEVICE_TABLE(of, adc_tm5_match_table);
+
+static struct platform_driver adc_tm5_driver = {
+	.driver = {
+		.name = "qcom-spmi-adc-tm5",
+		.of_match_table = adc_tm5_match_table,
+	},
+	.probe = adc_tm5_probe,
+};
+module_platform_driver(adc_tm5_driver);
+
+MODULE_DESCRIPTION("SPMI PMIC Thermal Monitor ADC driver");
+MODULE_LICENSE("GPL v2");
-- 
2.28.0


^ permalink raw reply related	[flat|nested] 15+ messages in thread

* [PATCH v5 8/9] arm64: dts: qcom: pm8150x: add definitions for adc-tm5 part
  2020-09-14 15:48 [PATCH v5 0/9] qcom: pm8150: add support for thermal monitoring Dmitry Baryshkov
                   ` (6 preceding siblings ...)
  2020-09-14 15:48 ` [PATCH v5 7/9] thermal: qcom: add support for adc-tm5 PMIC thermal monitor Dmitry Baryshkov
@ 2020-09-14 15:48 ` Dmitry Baryshkov
  2020-09-14 15:48 ` [PATCH v5 9/9] arm64: dts: sm8250-mtp: add thermal zones using pmic's adc-tm5 Dmitry Baryshkov
  8 siblings, 0 replies; 15+ messages in thread
From: Dmitry Baryshkov @ 2020-09-14 15:48 UTC (permalink / raw)
  To: Andy Gross, Bjorn Andersson, Rob Herring, Zhang Rui,
	Daniel Lezcano, Amit Kucheria, Jonathan Cameron, Hartmut Knaack,
	Lars-Peter Clausen, Peter Meerwald-Stadler
  Cc: linux-arm-msm, linux-pm, devicetree, linux-iio, Manivannan Sadhasivam

Define adc-tm5 thermal monitoring part. Individual channes and thermal
zones are to be configured in per-device dts files.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Acked-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
---
 arch/arm64/boot/dts/qcom/pm8150.dtsi  | 10 ++++++++++
 arch/arm64/boot/dts/qcom/pm8150b.dtsi | 10 ++++++++++
 arch/arm64/boot/dts/qcom/pm8150l.dtsi | 10 ++++++++++
 3 files changed, 30 insertions(+)

diff --git a/arch/arm64/boot/dts/qcom/pm8150.dtsi b/arch/arm64/boot/dts/qcom/pm8150.dtsi
index 1b6406927509..b1b518c6a2c9 100644
--- a/arch/arm64/boot/dts/qcom/pm8150.dtsi
+++ b/arch/arm64/boot/dts/qcom/pm8150.dtsi
@@ -97,6 +97,16 @@ die-temp@6 {
 			};
 		};
 
+		pm8150_adc_tm: adc-tm@3500 {
+			compatible = "qcom,spmi-adc-tm5";
+			reg = <0x3500>;
+			interrupts = <0x0 0x35 0x0 IRQ_TYPE_EDGE_RISING>;
+			#thermal-sensor-cells = <1>;
+			#address-cells = <1>;
+			#size-cells = <0>;
+			status = "disabled";
+		};
+
 		rtc@6000 {
 			compatible = "qcom,pm8941-rtc";
 			reg = <0x6000>;
diff --git a/arch/arm64/boot/dts/qcom/pm8150b.dtsi b/arch/arm64/boot/dts/qcom/pm8150b.dtsi
index e112e8876db6..8e2f3250c914 100644
--- a/arch/arm64/boot/dts/qcom/pm8150b.dtsi
+++ b/arch/arm64/boot/dts/qcom/pm8150b.dtsi
@@ -95,6 +95,16 @@ chg-temp@9 {
 			};
 		};
 
+		pm8150b_adc_tm: adc-tm@3500 {
+			compatible = "qcom,spmi-adc-tm5";
+			reg = <0x3500>;
+			interrupts = <0x2 0x35 0x0 IRQ_TYPE_EDGE_RISING>;
+			#thermal-sensor-cells = <1>;
+			#address-cells = <1>;
+			#size-cells = <0>;
+			status = "disabled";
+		};
+
 		pm8150b_gpios: gpio@c000 {
 			compatible = "qcom,pm8150b-gpio";
 			reg = <0xc000>;
diff --git a/arch/arm64/boot/dts/qcom/pm8150l.dtsi b/arch/arm64/boot/dts/qcom/pm8150l.dtsi
index 62139538b7d9..9f214ceec2b7 100644
--- a/arch/arm64/boot/dts/qcom/pm8150l.dtsi
+++ b/arch/arm64/boot/dts/qcom/pm8150l.dtsi
@@ -89,6 +89,16 @@ die-temp@6 {
 			};
 		};
 
+		pm8150l_adc_tm: adc-tm@3500 {
+			compatible = "qcom,spmi-adc-tm5";
+			reg = <0x3500>;
+			interrupts = <0x4 0x35 0x0 IRQ_TYPE_EDGE_RISING>;
+			#thermal-sensor-cells = <1>;
+			#address-cells = <1>;
+			#size-cells = <0>;
+			status = "disabled";
+		};
+
 		pm8150l_gpios: gpio@c000 {
 			compatible = "qcom,pm8150l-gpio";
 			reg = <0xc000>;
-- 
2.28.0


^ permalink raw reply related	[flat|nested] 15+ messages in thread

* [PATCH v5 9/9] arm64: dts: sm8250-mtp: add thermal zones using pmic's adc-tm5
  2020-09-14 15:48 [PATCH v5 0/9] qcom: pm8150: add support for thermal monitoring Dmitry Baryshkov
                   ` (7 preceding siblings ...)
  2020-09-14 15:48 ` [PATCH v5 8/9] arm64: dts: qcom: pm8150x: add definitions for adc-tm5 part Dmitry Baryshkov
@ 2020-09-14 15:48 ` Dmitry Baryshkov
  8 siblings, 0 replies; 15+ messages in thread
From: Dmitry Baryshkov @ 2020-09-14 15:48 UTC (permalink / raw)
  To: Andy Gross, Bjorn Andersson, Rob Herring, Zhang Rui,
	Daniel Lezcano, Amit Kucheria, Jonathan Cameron, Hartmut Knaack,
	Lars-Peter Clausen, Peter Meerwald-Stadler
  Cc: linux-arm-msm, linux-pm, devicetree, linux-iio, Manivannan Sadhasivam

Port thermal zones definitions from msm-4.19 tree. Enable and add
channel configuration to PMIC's ADC-TM definitions. Declare thermal
zones and respective trip points.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
---
 arch/arm64/boot/dts/qcom/sm8250-mtp.dts | 212 ++++++++++++++++++++++++
 1 file changed, 212 insertions(+)

diff --git a/arch/arm64/boot/dts/qcom/sm8250-mtp.dts b/arch/arm64/boot/dts/qcom/sm8250-mtp.dts
index 6e2f7ae1d621..d06207ac5593 100644
--- a/arch/arm64/boot/dts/qcom/sm8250-mtp.dts
+++ b/arch/arm64/boot/dts/qcom/sm8250-mtp.dts
@@ -24,6 +24,104 @@ chosen {
 		stdout-path = "serial0:115200n8";
 	};
 
+	thermal-zones {
+		xo-therm {
+			polling-delay-passive = <0>;
+			polling-delay = <0>;
+			thermal-sensors = <&pm8150_adc_tm 0>;
+			trips {
+				active-config0 {
+					temperature = <125000>;
+					hysteresis = <1000>;
+					type = "passive";
+				};
+			};
+		};
+
+		skin-therm {
+			polling-delay-passive = <0>;
+			polling-delay = <0>;
+			thermal-sensors = <&pm8150_adc_tm 1>;
+			trips {
+				active-config0 {
+					temperature = <125000>;
+					hysteresis = <1000>;
+					type = "passive";
+				};
+			};
+		};
+
+		mmw-pa1 {
+			polling-delay-passive = <0>;
+			polling-delay = <0>;
+			thermal-sensors = <&pm8150_adc_tm 2>;
+
+			trips {
+				active-config0 {
+					temperature = <125000>;
+					hysteresis = <1000>;
+					type = "passive";
+				};
+			};
+		};
+
+		conn-therm {
+			polling-delay-passive = <0>;
+			polling-delay = <0>;
+			thermal-sensors = <&pm8150b_adc_tm 0>;
+
+			trips {
+				active-config0 {
+					temperature = <125000>;
+					hysteresis = <1000>;
+					type = "passive";
+				};
+			};
+		};
+
+		camera-therm {
+			polling-delay-passive = <0>;
+			polling-delay = <0>;
+			thermal-sensors = <&pm8150l_adc_tm 0>;
+
+			trips {
+				active-config0 {
+					temperature = <125000>;
+					hysteresis = <1000>;
+					type = "passive";
+				};
+			};
+		};
+
+		skin-msm-therm {
+			polling-delay-passive = <0>;
+			polling-delay = <0>;
+			thermal-sensors = <&pm8150l_adc_tm 1>;
+
+			trips {
+				active-config0 {
+					temperature = <125000>;
+					hysteresis = <1000>;
+					type = "passive";
+				};
+			};
+		};
+
+		mmw-pa2 {
+			polling-delay-passive = <0>;
+			polling-delay = <0>;
+			thermal-sensors = <&pm8150l_adc_tm 2>;
+
+			trips {
+				active-config0 {
+					temperature = <125000>;
+					hysteresis = <1000>;
+					type = "passive";
+				};
+			};
+		};
+	};
+
 	vph_pwr: vph-pwr-regulator {
 		compatible = "regulator-fixed";
 		regulator-name = "vph_pwr";
@@ -358,6 +456,120 @@ &cdsp {
 	firmware-name = "qcom/sm8250/cdsp.mbn";
 };
 
+&pm8150_adc {
+	xo-therm@4c {
+		reg = <ADC5_XO_THERM_100K_PU>;
+		qcom,ratiometric;
+		qcom,hw-settle-time = <200>;
+	};
+
+	skin-therm@4d {
+		reg = <ADC5_AMUX_THM1_100K_PU>;
+		qcom,ratiometric;
+		qcom,hw-settle-time = <200>;
+	};
+
+	pa-therm1@4e {
+		reg = <ADC5_AMUX_THM2_100K_PU>;
+		qcom,ratiometric;
+		qcom,hw-settle-time = <200>;
+	};
+};
+
+&pm8150b_adc {
+	conn-therm@4f {
+		reg = <ADC5_AMUX_THM3_100K_PU>;
+		qcom,ratiometric;
+		qcom,hw-settle-time = <200>;
+	};
+};
+
+&pm8150l_adc {
+	camera-flash-therm@4d {
+		reg = <ADC5_AMUX_THM1_100K_PU>;
+		qcom,ratiometric;
+		qcom,hw-settle-time = <200>;
+	};
+
+	skin-msm-therm@4e {
+		reg = <ADC5_AMUX_THM2_100K_PU>;
+		qcom,ratiometric;
+		qcom,hw-settle-time = <200>;
+	};
+
+	pa-therm2@4f {
+		reg = <ADC5_AMUX_THM3_100K_PU>;
+		qcom,ratiometric;
+		qcom,hw-settle-time = <200>;
+	};
+};
+
+&pm8150_adc_tm {
+	status = "okay";
+
+	xo-therm@0 {
+		reg = <0>;
+		io-channels = <&pm8150_adc ADC5_XO_THERM_100K_PU>;
+		qcom,adc-channel = <ADC5_XO_THERM_100K_PU>;
+		qcom,ratiometric;
+		qcom,hw-settle-time = <200>;
+	};
+
+	skin-therm@1 {
+		reg = <1>;
+		io-channels = <&pm8150_adc ADC5_AMUX_THM1_100K_PU>;
+		qcom,adc-channel = <ADC5_AMUX_THM1_100K_PU>;
+		qcom,ratiometric;
+		qcom,hw-settle-time = <200>;
+	};
+
+	pa-therm1@2 {
+		reg = <2>;
+		io-channels = <&pm8150_adc ADC5_AMUX_THM2_100K_PU>;
+		qcom,ratiometric;
+		qcom,hw-settle-time = <200>;
+	};
+};
+
+&pm8150b_adc_tm {
+	status = "okay";
+
+	conn-therm@0 {
+		reg = <0>;
+		io-channels = <&pm8150b_adc ADC5_AMUX_THM3_100K_PU>;
+		qcom,adc-channel = <ADC5_AMUX_THM3_100K_PU>;
+		qcom,ratiometric;
+		qcom,hw-settle-time = <200>;
+	};
+};
+
+&pm8150l_adc_tm {
+	status = "okay";
+
+	camera-flash-therm@0 {
+		reg = <0>;
+		io-channels = <&pm8150l_adc ADC5_AMUX_THM1_100K_PU>;
+		qcom,adc-channel = <ADC5_AMUX_THM1_100K_PU>;
+		qcom,ratiometric;
+		qcom,hw-settle-time = <200>;
+	};
+
+	skin-msm-therm@1 {
+		reg = <1>;
+		io-channels = <&pm8150l_adc ADC5_AMUX_THM2_100K_PU>;
+		qcom,adc-channel = <ADC5_AMUX_THM2_100K_PU>;
+		qcom,ratiometric;
+		qcom,hw-settle-time = <200>;
+	};
+
+	pa-therm2@2 {
+		reg = <2>;
+		io-channels = <&pm8150l_adc ADC5_AMUX_THM3_100K_PU>;
+		qcom,ratiometric;
+		qcom,hw-settle-time = <200>;
+	};
+};
+
 &qupv3_id_1 {
 	status = "okay";
 };
-- 
2.28.0


^ permalink raw reply related	[flat|nested] 15+ messages in thread

* Re: [PATCH v5 1/9] dt-bindings: thermal: qcom: add adc-thermal monitor bindings
  2020-09-14 15:48 ` [PATCH v5 1/9] dt-bindings: thermal: qcom: add adc-thermal monitor bindings Dmitry Baryshkov
@ 2020-09-22 23:40   ` Rob Herring
  2020-09-23  9:07     ` Dmitry Baryshkov
  0 siblings, 1 reply; 15+ messages in thread
From: Rob Herring @ 2020-09-22 23:40 UTC (permalink / raw)
  To: Dmitry Baryshkov
  Cc: Andy Gross, Bjorn Andersson, Zhang Rui, Daniel Lezcano,
	Amit Kucheria, Jonathan Cameron, Hartmut Knaack,
	Lars-Peter Clausen, Peter Meerwald-Stadler, linux-arm-msm,
	linux-pm, devicetree, linux-iio, Manivannan Sadhasivam,
	Jonathan Cameron

On Mon, Sep 14, 2020 at 06:48:01PM +0300, Dmitry Baryshkov wrote:
> Add bindings for thermal monitor, part of Qualcomm PMIC5 chips. It is a
> close counterpart of VADC part of those PMICs.
> 
> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> ---
>  .../bindings/thermal/qcom-spmi-adc-tm5.yaml   | 151 ++++++++++++++++++
>  1 file changed, 151 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml
> 
> diff --git a/Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml b/Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml
> new file mode 100644
> index 000000000000..432a65839b89
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml
> @@ -0,0 +1,151 @@
> +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
> +%YAML 1.2
> +---
> +$id: http://devicetree.org/schemas/thermal/qcom-spmi-adc-tm5.yaml#
> +$schema: http://devicetree.org/meta-schemas/core.yaml#
> +
> +title: Qualcomm's SPMI PMIC ADC Thermal Monitoring
> +maintainers:
> +  - Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> +
> +properties:
> +  compatible:
> +    const: qcom,spmi-adc-tm5
> +
> +  reg:
> +    maxItems: 1
> +
> +  interrupts:
> +    maxItems: 1
> +
> +  "#thermal-sensor-cells":
> +    const: 1
> +    description:
> +      Number of cells required to uniquely identify the thermal sensors. Since
> +      we have multiple sensors this is set to 1
> +
> +  "#address-cells":
> +    const: 1
> +
> +  "#size-cells":
> +    const: 0
> +
> +  qcom,avg-samples:
> +    $ref: /schemas/types.yaml#/definitions/uint32
> +    description: Number of samples to be used for measurement.
> +    enum:
> +      - 1
> +      - 2
> +      - 4
> +      - 8
> +      - 16
> +    default: 1
> +
> +  qcom,decimation:
> +    $ref: /schemas/types.yaml#/definitions/uint32
> +    description: This parameter is used to decrease ADC sampling rate.
> +            Quicker measurements can be made by reducing decimation ratio.
> +    enum:
> +      - 250
> +      - 420
> +      - 840
> +    default: 840
> +
> +patternProperties:
> +  "^([-a-z0-9]*)@[0-9]+$":

Less than 10 as unit-addresses are hex?

> +    type: object
> +    description:
> +      Represent one thermal sensor.
> +
> +    properties:
> +      reg:
> +        description: Specify the sensor channel.
> +        maxItems: 1

You need a range of values here.

> +
> +      io-channels:
> +        description:
> +          From common IIO binding. Used to pipe PMIC ADC channel to thermal monitor
> +
> +      qcom,adc-channel:
> +        $ref: /schemas/types.yaml#/definitions/uint32
> +        description: Corresponding ADC channel ID.

Why is this not a cell in io-channels?

> +
> +      qcom,ratiometric:
> +        $ref: /schemas/types.yaml#/definitions/flag
> +        description:
> +          Channel calibration type.
> +          If this property is specified VADC will use the VDD reference
> +          (1.875V) and GND for channel calibration. If property is not found,
> +          channel will be calibrated with 0V and 1.25V reference channels,
> +          also known as absolute calibration.
> +
> +      qcom,hw-settle-time:
> +        $ref: /schemas/types.yaml#/definitions/uint32
> +        description: Time between AMUX getting configured and the ADC starting conversion.

Time values should have a unit suffix. Seems like a commmon ADC 
property...

> +
> +      qcom,pre-scaling:
> +        $ref: /schemas/types.yaml#/definitions/uint32-array
> +        description: Used for scaling the channel input signal before the
> +          signal is fed to VADC. See qcom,spi-vadc specification for the list
> +          of possible values.

I'd rather not. Need the values here to validate a DT.

> +        minItems: 2
> +        maxItems: 2
> +
> +    required:
> +      - reg
> +      - qcom,adc-channel
> +
> +    additionalProperties:
> +      false
> +
> +required:
> +  - compatible
> +  - reg
> +  - interrupts
> +  - "#address-cells"
> +  - "#size-cells"
> +  - "#thermal-sensor-cells"
> +
> +additionalProperties: false
> +
> +examples:
> +  - |
> +    #include <dt-bindings/iio/qcom,spmi-vadc.h>
> +    #include <dt-bindings/interrupt-controller/irq.h>
> +    spmi_bus {
> +        #address-cells = <1>;
> +        #size-cells = <0>;
> +        pm8150b_adc: adc@3100 {
> +            reg = <0x3100>;
> +            compatible = "qcom,spmi-adc5";
> +            #address-cells = <1>;
> +            #size-cells = <0>;
> +            #io-channel-cells = <1>;
> +            io-channel-ranges;
> +
> +            /* Other propreties are omitted */
> +            conn-therm@4f {
> +                reg = <ADC5_AMUX_THM3_100K_PU>;
> +                qcom,ratiometric;
> +                qcom,hw-settle-time = <200>;
> +            };
> +        };
> +
> +        pm8150b_adc_tm: adc-tm@3500 {
> +            compatible = "qcom,spmi-adc-tm5";
> +            reg = <0x3500>;
> +            interrupts = <0x2 0x35 0x0 IRQ_TYPE_EDGE_RISING>;
> +            #thermal-sensor-cells = <1>;
> +            #address-cells = <1>;
> +            #size-cells = <0>;
> +
> +            conn-therm@0 {
> +                reg = <0>;
> +                io-channels = <&pm8150b_adc ADC5_AMUX_THM3_100K_PU>;
> +                qcom,adc-channel = <ADC5_AMUX_THM3_100K_PU>;
> +                qcom,ratiometric;
> +                qcom,hw-settle-time = <200>;
> +            };
> +        };
> +    };
> +...
> -- 
> 2.28.0
> 

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: [PATCH v5 1/9] dt-bindings: thermal: qcom: add adc-thermal monitor bindings
  2020-09-22 23:40   ` Rob Herring
@ 2020-09-23  9:07     ` Dmitry Baryshkov
  2020-09-23 14:00       ` Rob Herring
  0 siblings, 1 reply; 15+ messages in thread
From: Dmitry Baryshkov @ 2020-09-23  9:07 UTC (permalink / raw)
  To: Rob Herring
  Cc: Andy Gross, Bjorn Andersson, Zhang Rui, Daniel Lezcano,
	Amit Kucheria, Jonathan Cameron, Hartmut Knaack,
	Lars-Peter Clausen, Peter Meerwald-Stadler, linux-arm-msm,
	linux-pm, devicetree, linux-iio, Manivannan Sadhasivam,
	Jonathan Cameron

On 23/09/2020 02:40, Rob Herring wrote:
> On Mon, Sep 14, 2020 at 06:48:01PM +0300, Dmitry Baryshkov wrote:
>> Add bindings for thermal monitor, part of Qualcomm PMIC5 chips. It is a
>> close counterpart of VADC part of those PMICs.
>>
>> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
>> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
>> ---
>>   .../bindings/thermal/qcom-spmi-adc-tm5.yaml   | 151 ++++++++++++++++++
>>   1 file changed, 151 insertions(+)
>>   create mode 100644 Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml
>>
>> diff --git a/Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml b/Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml
>> new file mode 100644
>> index 000000000000..432a65839b89
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml
>> @@ -0,0 +1,151 @@
>> +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
>> +%YAML 1.2
>> +---
>> +$id: http://devicetree.org/schemas/thermal/qcom-spmi-adc-tm5.yaml#
>> +$schema: http://devicetree.org/meta-schemas/core.yaml#
>> +
>> +title: Qualcomm's SPMI PMIC ADC Thermal Monitoring
>> +maintainers:
>> +  - Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
>> +
>> +properties:
>> +  compatible:
>> +    const: qcom,spmi-adc-tm5
>> +
>> +  reg:
>> +    maxItems: 1
>> +
>> +  interrupts:
>> +    maxItems: 1
>> +
>> +  "#thermal-sensor-cells":
>> +    const: 1
>> +    description:
>> +      Number of cells required to uniquely identify the thermal sensors. Since
>> +      we have multiple sensors this is set to 1
>> +
>> +  "#address-cells":
>> +    const: 1
>> +
>> +  "#size-cells":
>> +    const: 0
>> +
>> +  qcom,avg-samples:
>> +    $ref: /schemas/types.yaml#/definitions/uint32
>> +    description: Number of samples to be used for measurement.
>> +    enum:
>> +      - 1
>> +      - 2
>> +      - 4
>> +      - 8
>> +      - 16
>> +    default: 1
>> +
>> +  qcom,decimation:
>> +    $ref: /schemas/types.yaml#/definitions/uint32
>> +    description: This parameter is used to decrease ADC sampling rate.
>> +            Quicker measurements can be made by reducing decimation ratio.
>> +    enum:
>> +      - 250
>> +      - 420
>> +      - 840
>> +    default: 840
>> +
>> +patternProperties:
>> +  "^([-a-z0-9]*)@[0-9]+$":
> 
> Less than 10 as unit-addresses are hex?

8 channels at max currently. I'll fix to use hex though.

> 
>> +    type: object
>> +    description:
>> +      Represent one thermal sensor.
>> +
>> +    properties:
>> +      reg:
>> +        description: Specify the sensor channel.
>> +        maxItems: 1
> 
> You need a range of values here.

ok.

> 
>> +
>> +      io-channels:
>> +        description:
>> +          From common IIO binding. Used to pipe PMIC ADC channel to thermal monitor
>> +
>> +      qcom,adc-channel:
>> +        $ref: /schemas/types.yaml#/definitions/uint32
>> +        description: Corresponding ADC channel ID.
> 
> Why is this not a cell in io-channels?


Do you mean parsing a cell from io-channels rather than specifying it 
again? Sounds like a good idea.

> 
>> +
>> +      qcom,ratiometric:
>> +        $ref: /schemas/types.yaml#/definitions/flag
>> +        description:
>> +          Channel calibration type.
>> +          If this property is specified VADC will use the VDD reference
>> +          (1.875V) and GND for channel calibration. If property is not found,
>> +          channel will be calibrated with 0V and 1.25V reference channels,
>> +          also known as absolute calibration.
>> +
>> +      qcom,hw-settle-time:
>> +        $ref: /schemas/types.yaml#/definitions/uint32
>> +        description: Time between AMUX getting configured and the ADC starting conversion.
> 
> Time values should have a unit suffix. Seems like a commmon ADC
> property...

Could you please be more specific here? Would you like for me to just 
specify the unit in the description?

> 
>> +
>> +      qcom,pre-scaling:
>> +        $ref: /schemas/types.yaml#/definitions/uint32-array
>> +        description: Used for scaling the channel input signal before the
>> +          signal is fed to VADC. See qcom,spi-vadc specification for the list
>> +          of possible values.
> 
> I'd rather not. Need the values here to validate a DT.

OK

> 
>> +        minItems: 2
>> +        maxItems: 2
>> +
>> +    required:
>> +      - reg
>> +      - qcom,adc-channel
>> +
>> +    additionalProperties:
>> +      false
>> +
>> +required:
>> +  - compatible
>> +  - reg
>> +  - interrupts
>> +  - "#address-cells"
>> +  - "#size-cells"
>> +  - "#thermal-sensor-cells"
>> +
>> +additionalProperties: false
>> +
>> +examples:
>> +  - |
>> +    #include <dt-bindings/iio/qcom,spmi-vadc.h>
>> +    #include <dt-bindings/interrupt-controller/irq.h>
>> +    spmi_bus {
>> +        #address-cells = <1>;
>> +        #size-cells = <0>;
>> +        pm8150b_adc: adc@3100 {
>> +            reg = <0x3100>;
>> +            compatible = "qcom,spmi-adc5";
>> +            #address-cells = <1>;
>> +            #size-cells = <0>;
>> +            #io-channel-cells = <1>;
>> +            io-channel-ranges;
>> +
>> +            /* Other propreties are omitted */
>> +            conn-therm@4f {
>> +                reg = <ADC5_AMUX_THM3_100K_PU>;
>> +                qcom,ratiometric;
>> +                qcom,hw-settle-time = <200>;
>> +            };
>> +        };
>> +
>> +        pm8150b_adc_tm: adc-tm@3500 {
>> +            compatible = "qcom,spmi-adc-tm5";
>> +            reg = <0x3500>;
>> +            interrupts = <0x2 0x35 0x0 IRQ_TYPE_EDGE_RISING>;
>> +            #thermal-sensor-cells = <1>;
>> +            #address-cells = <1>;
>> +            #size-cells = <0>;
>> +
>> +            conn-therm@0 {
>> +                reg = <0>;
>> +                io-channels = <&pm8150b_adc ADC5_AMUX_THM3_100K_PU>;
>> +                qcom,adc-channel = <ADC5_AMUX_THM3_100K_PU>;
>> +                qcom,ratiometric;
>> +                qcom,hw-settle-time = <200>;
>> +            };
>> +        };
>> +    };
>> +...
>> -- 
>> 2.28.0
>>


-- 
With best wishes
Dmitry

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: [PATCH v5 1/9] dt-bindings: thermal: qcom: add adc-thermal monitor bindings
  2020-09-23  9:07     ` Dmitry Baryshkov
@ 2020-09-23 14:00       ` Rob Herring
  2020-09-23 15:22         ` Jonathan Cameron
  0 siblings, 1 reply; 15+ messages in thread
From: Rob Herring @ 2020-09-23 14:00 UTC (permalink / raw)
  To: Dmitry Baryshkov
  Cc: Andy Gross, Bjorn Andersson, Zhang Rui, Daniel Lezcano,
	Amit Kucheria, Jonathan Cameron, Hartmut Knaack,
	Lars-Peter Clausen, Peter Meerwald-Stadler, linux-arm-msm,
	open list:THERMAL, devicetree,
	open list:IIO SUBSYSTEM AND DRIVERS, Manivannan Sadhasivam,
	Jonathan Cameron

On Wed, Sep 23, 2020 at 3:07 AM Dmitry Baryshkov
<dmitry.baryshkov@linaro.org> wrote:
>
> On 23/09/2020 02:40, Rob Herring wrote:
> > On Mon, Sep 14, 2020 at 06:48:01PM +0300, Dmitry Baryshkov wrote:
> >> Add bindings for thermal monitor, part of Qualcomm PMIC5 chips. It is a
> >> close counterpart of VADC part of those PMICs.
> >>
> >> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> >> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> >> ---
> >>   .../bindings/thermal/qcom-spmi-adc-tm5.yaml   | 151 ++++++++++++++++++
> >>   1 file changed, 151 insertions(+)
> >>   create mode 100644 Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml
> >>
> >> diff --git a/Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml b/Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml
> >> new file mode 100644
> >> index 000000000000..432a65839b89
> >> --- /dev/null
> >> +++ b/Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml
> >> @@ -0,0 +1,151 @@
> >> +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
> >> +%YAML 1.2
> >> +---
> >> +$id: http://devicetree.org/schemas/thermal/qcom-spmi-adc-tm5.yaml#
> >> +$schema: http://devicetree.org/meta-schemas/core.yaml#
> >> +
> >> +title: Qualcomm's SPMI PMIC ADC Thermal Monitoring
> >> +maintainers:
> >> +  - Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> >> +
> >> +properties:
> >> +  compatible:
> >> +    const: qcom,spmi-adc-tm5
> >> +
> >> +  reg:
> >> +    maxItems: 1
> >> +
> >> +  interrupts:
> >> +    maxItems: 1
> >> +
> >> +  "#thermal-sensor-cells":
> >> +    const: 1
> >> +    description:
> >> +      Number of cells required to uniquely identify the thermal sensors. Since
> >> +      we have multiple sensors this is set to 1
> >> +
> >> +  "#address-cells":
> >> +    const: 1
> >> +
> >> +  "#size-cells":
> >> +    const: 0
> >> +
> >> +  qcom,avg-samples:
> >> +    $ref: /schemas/types.yaml#/definitions/uint32
> >> +    description: Number of samples to be used for measurement.
> >> +    enum:
> >> +      - 1
> >> +      - 2
> >> +      - 4
> >> +      - 8
> >> +      - 16
> >> +    default: 1
> >> +
> >> +  qcom,decimation:
> >> +    $ref: /schemas/types.yaml#/definitions/uint32
> >> +    description: This parameter is used to decrease ADC sampling rate.
> >> +            Quicker measurements can be made by reducing decimation ratio.
> >> +    enum:
> >> +      - 250
> >> +      - 420
> >> +      - 840
> >> +    default: 840
> >> +
> >> +patternProperties:
> >> +  "^([-a-z0-9]*)@[0-9]+$":
> >
> > Less than 10 as unit-addresses are hex?
>
> 8 channels at max currently. I'll fix to use hex though.

Then it should be @[0-7]$

> >> +    type: object
> >> +    description:
> >> +      Represent one thermal sensor.
> >> +
> >> +    properties:
> >> +      reg:
> >> +        description: Specify the sensor channel.
> >> +        maxItems: 1
> >
> > You need a range of values here.
>
> ok.
>
> >
> >> +
> >> +      io-channels:
> >> +        description:
> >> +          From common IIO binding. Used to pipe PMIC ADC channel to thermal monitor
> >> +
> >> +      qcom,adc-channel:
> >> +        $ref: /schemas/types.yaml#/definitions/uint32
> >> +        description: Corresponding ADC channel ID.
> >
> > Why is this not a cell in io-channels?
>
>
> Do you mean parsing a cell from io-channels rather than specifying it
> again? Sounds like a good idea.

Yes.

> >> +      qcom,ratiometric:
> >> +        $ref: /schemas/types.yaml#/definitions/flag
> >> +        description:
> >> +          Channel calibration type.
> >> +          If this property is specified VADC will use the VDD reference
> >> +          (1.875V) and GND for channel calibration. If property is not found,
> >> +          channel will be calibrated with 0V and 1.25V reference channels,
> >> +          also known as absolute calibration.
> >> +
> >> +      qcom,hw-settle-time:
> >> +        $ref: /schemas/types.yaml#/definitions/uint32
> >> +        description: Time between AMUX getting configured and the ADC starting conversion.
> >
> > Time values should have a unit suffix. Seems like a commmon ADC
> > property...
>
> Could you please be more specific here? Would you like for me to just
> specify the unit in the description?

More a question for Jonathan I guess as to whether this should be
common or not. Maybe we have something already. Settle or acquisition
time is a common thing for ADCs, right?

Properties with units need a suffix as defined in
.../bindings/property-units.txt.

Rob

^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: [PATCH v5 1/9] dt-bindings: thermal: qcom: add adc-thermal monitor bindings
  2020-09-23 14:00       ` Rob Herring
@ 2020-09-23 15:22         ` Jonathan Cameron
  0 siblings, 0 replies; 15+ messages in thread
From: Jonathan Cameron @ 2020-09-23 15:22 UTC (permalink / raw)
  To: Rob Herring
  Cc: Dmitry Baryshkov, Andy Gross, Bjorn Andersson, Zhang Rui,
	Daniel Lezcano, Amit Kucheria, Jonathan Cameron, Hartmut Knaack,
	Lars-Peter Clausen, Peter Meerwald-Stadler, linux-arm-msm,
	open list:THERMAL, devicetree,
	open list:IIO SUBSYSTEM AND DRIVERS, Manivannan Sadhasivam

On Wed, 23 Sep 2020 08:00:29 -0600
Rob Herring <robh@kernel.org> wrote:

> On Wed, Sep 23, 2020 at 3:07 AM Dmitry Baryshkov
> <dmitry.baryshkov@linaro.org> wrote:
> >
> > On 23/09/2020 02:40, Rob Herring wrote:  
> > > On Mon, Sep 14, 2020 at 06:48:01PM +0300, Dmitry Baryshkov wrote:  
> > >> Add bindings for thermal monitor, part of Qualcomm PMIC5 chips. It is a
> > >> close counterpart of VADC part of those PMICs.
> > >>
> > >> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> > >> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> > >> ---
> > >>   .../bindings/thermal/qcom-spmi-adc-tm5.yaml   | 151 ++++++++++++++++++
> > >>   1 file changed, 151 insertions(+)
> > >>   create mode 100644 Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml
> > >>
> > >> diff --git a/Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml b/Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml
> > >> new file mode 100644
> > >> index 000000000000..432a65839b89
> > >> --- /dev/null
> > >> +++ b/Documentation/devicetree/bindings/thermal/qcom-spmi-adc-tm5.yaml
> > >> @@ -0,0 +1,151 @@
> > >> +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
> > >> +%YAML 1.2
> > >> +---
> > >> +$id: http://devicetree.org/schemas/thermal/qcom-spmi-adc-tm5.yaml#
> > >> +$schema: http://devicetree.org/meta-schemas/core.yaml#
> > >> +
> > >> +title: Qualcomm's SPMI PMIC ADC Thermal Monitoring
> > >> +maintainers:
> > >> +  - Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> > >> +
> > >> +properties:
> > >> +  compatible:
> > >> +    const: qcom,spmi-adc-tm5
> > >> +
> > >> +  reg:
> > >> +    maxItems: 1
> > >> +
> > >> +  interrupts:
> > >> +    maxItems: 1
> > >> +
> > >> +  "#thermal-sensor-cells":
> > >> +    const: 1
> > >> +    description:
> > >> +      Number of cells required to uniquely identify the thermal sensors. Since
> > >> +      we have multiple sensors this is set to 1
> > >> +
> > >> +  "#address-cells":
> > >> +    const: 1
> > >> +
> > >> +  "#size-cells":
> > >> +    const: 0
> > >> +
> > >> +  qcom,avg-samples:
> > >> +    $ref: /schemas/types.yaml#/definitions/uint32
> > >> +    description: Number of samples to be used for measurement.
> > >> +    enum:
> > >> +      - 1
> > >> +      - 2
> > >> +      - 4
> > >> +      - 8
> > >> +      - 16
> > >> +    default: 1
> > >> +
> > >> +  qcom,decimation:
> > >> +    $ref: /schemas/types.yaml#/definitions/uint32
> > >> +    description: This parameter is used to decrease ADC sampling rate.
> > >> +            Quicker measurements can be made by reducing decimation ratio.
> > >> +    enum:
> > >> +      - 250
> > >> +      - 420
> > >> +      - 840
> > >> +    default: 840
> > >> +
> > >> +patternProperties:
> > >> +  "^([-a-z0-9]*)@[0-9]+$":  
> > >
> > > Less than 10 as unit-addresses are hex?  
> >
> > 8 channels at max currently. I'll fix to use hex though.  
> 
> Then it should be @[0-7]$
> 
> > >> +    type: object
> > >> +    description:
> > >> +      Represent one thermal sensor.
> > >> +
> > >> +    properties:
> > >> +      reg:
> > >> +        description: Specify the sensor channel.
> > >> +        maxItems: 1  
> > >
> > > You need a range of values here.  
> >
> > ok.
> >  
> > >  
> > >> +
> > >> +      io-channels:
> > >> +        description:
> > >> +          From common IIO binding. Used to pipe PMIC ADC channel to thermal monitor
> > >> +
> > >> +      qcom,adc-channel:
> > >> +        $ref: /schemas/types.yaml#/definitions/uint32
> > >> +        description: Corresponding ADC channel ID.  
> > >
> > > Why is this not a cell in io-channels?  
> >
> >
> > Do you mean parsing a cell from io-channels rather than specifying it
> > again? Sounds like a good idea.  
> 
> Yes.
> 
> > >> +      qcom,ratiometric:
> > >> +        $ref: /schemas/types.yaml#/definitions/flag
> > >> +        description:
> > >> +          Channel calibration type.
> > >> +          If this property is specified VADC will use the VDD reference
> > >> +          (1.875V) and GND for channel calibration. If property is not found,
> > >> +          channel will be calibrated with 0V and 1.25V reference channels,
> > >> +          also known as absolute calibration.
> > >> +
> > >> +      qcom,hw-settle-time:
> > >> +        $ref: /schemas/types.yaml#/definitions/uint32
> > >> +        description: Time between AMUX getting configured and the ADC starting conversion.  
> > >
> > > Time values should have a unit suffix. Seems like a commmon ADC
> > > property...  
> >
> > Could you please be more specific here? Would you like for me to just
> > specify the unit in the description?  
> 
> More a question for Jonathan I guess as to whether this should be
> common or not. Maybe we have something already. Settle or acquisition
> time is a common thing for ADCs, right?

It's not common in my experience, but not unheard of.
Only cases currently supporting controlling it explicitly in the
kernel that I can spot, are currently all qcom parts
(I'd guess they are all using the same IP underneath)

I think it is usually it's a case of building your analog
circuitry to match the spec of your ADC / MUX rather than
relaxing that spec by adding a delay.

It's a property with obvious enough meaning though that I
wouldn't have a problem with it being a generic property even
it is one that doesn't actually get used much.

> 
> Properties with units need a suffix as defined in
> .../bindings/property-units.txt.

We have a bunch of legacy bindings that don't have units but
all new ones certainly should!

Thanks,

Jonathan

> 
> Rob



^ permalink raw reply	[flat|nested] 15+ messages in thread

* Re: [PATCH v5 7/9] thermal: qcom: add support for adc-tm5 PMIC thermal monitor
  2020-09-14 15:48 ` [PATCH v5 7/9] thermal: qcom: add support for adc-tm5 PMIC thermal monitor Dmitry Baryshkov
@ 2020-09-25 11:42   ` Jishnu Prakash
  0 siblings, 0 replies; 15+ messages in thread
From: Jishnu Prakash @ 2020-09-25 11:42 UTC (permalink / raw)
  To: Dmitry Baryshkov, Andy Gross, Bjorn Andersson, Rob Herring,
	Zhang Rui, Daniel Lezcano, Amit Kucheria, Jonathan Cameron,
	Hartmut Knaack, Lars-Peter Clausen, Peter Meerwald-Stadler
  Cc: linux-arm-msm, linux-pm, devicetree, linux-iio,
	Manivannan Sadhasivam, Jonathan Cameron

Hi Dmitry,

Recently I was testing the ADC_TM driver added with these changes on 
SC7180 and I could see that this patch needs a few changes, which I'll 
mention below.

On 9/14/2020 9:18 PM, Dmitry Baryshkov wrote:
> Add support for Thermal Monitoring part of PMIC5. This part is closely
> coupled with ADC, using it's channels directly. ADC-TM support
> generating interrupts on ADC value crossing low or high voltage bounds,
> which is used to support thermal trip points.
>
> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> ---
>   drivers/iio/adc/qcom-vadc-common.c       |  62 +++
>   drivers/iio/adc/qcom-vadc-common.h       |   3 +
>   drivers/thermal/qcom/Kconfig             |  11 +
>   drivers/thermal/qcom/Makefile            |   1 +
>   drivers/thermal/qcom/qcom-spmi-adc-tm5.c | 583 +++++++++++++++++++++++
>   5 files changed, 660 insertions(+)
>   create mode 100644 drivers/thermal/qcom/qcom-spmi-adc-tm5.c
>
> diff --git a/drivers/iio/adc/qcom-vadc-common.c b/drivers/iio/adc/qcom-vadc-common.c
> index 40d77b3af1bb..e58e393b8713 100644
> --- a/drivers/iio/adc/qcom-vadc-common.c
> +++ b/drivers/iio/adc/qcom-vadc-common.c
> @@ -377,6 +377,42 @@ static int qcom_vadc_map_voltage_temp(const struct vadc_map_pt *pts,
>   	return 0;
>   }
>   


> +static irqreturn_t adc_tm5_isr(int irq, void *data)
> +{
> +	struct adc_tm5_chip *chip = data;
> +	u8 status_low, status_high, ctl;
> +	int ret = 0, i = 0;
> +
> +	ret = adc_tm5_read(chip, ADC_TM5_STATUS_LOW, &status_low, 1);
> +	if (ret) {
> +		dev_err(chip->dev, "read status low failed with %d\n", ret);
> +		return IRQ_HANDLED;
> +	}
> +
> +	ret = adc_tm5_read(chip, ADC_TM5_STATUS_HIGH, &status_high, 1);
> +	if (ret) {
> +		dev_err(chip->dev, "read status high failed with %d\n", ret);
> +		return IRQ_HANDLED;
> +	}
> +
> +	for (i = 0; i < chip->nchannels; i++) {
> +		bool upper_set = false, lower_set = false;
> +		unsigned int ch = chip->channels[i].channel;
> +
> +		if (!chip->channels[i].tzd) {
> +			dev_err_once(chip->dev, "thermal device not found\n");
> +			continue;
> +		}
> +
> +		ret = adc_tm5_read(chip, ADC_TM5_M_EN(ch), &ctl, 1);
> +
> +		if (ret) {
> +			dev_err(chip->dev, "ctl read failed with %d\n", ret);
> +			continue;
> +		}
> +
> +		lower_set = (status_low & BIT(ch)) &&
> +			(ctl & ADC_TM5_M_MEAS_EN) &&
> +			(ctl & ADC_TM5_M_LOW_THR_INT_EN);
> +
> +		upper_set = (status_high & BIT(ch)) &&
> +			(ctl & ADC_TM5_M_MEAS_EN) &&
> +			(ctl & ADC_TM5_M_HIGH_THR_INT_EN);
> +
> +		if (upper_set || lower_set)
> +			thermal_zone_device_update(chip->channels[i].tzd,
> +						   THERMAL_EVENT_UNSPECIFIED);

When using thermal_zone_device_update here, it internally calls 
tz->ops->get_temp, which maps to adc_tm5_get_temp defined just below. 
This in turn calls iio_read_channel_processed, which internally calls a 
mutex and this results in a mutex being called from atomic context.

To avoid this, the interrupt should be requested as a threaded IRQ.

> +	}
> +
> +	return IRQ_HANDLED;
> +}
> +
> +static int adc_tm5_get_temp(void *data, int *temp)
> +{
> +	struct adc_tm5_channel *channel = data;
> +	int ret, milli_celsius;
> +
> +	if (!channel || !channel->iio)
> +		return -EINVAL;
> +
> +	ret = iio_read_channel_processed(channel->iio, &milli_celsius);
> +	if (ret < 0)
> +		return ret;
> +
> +	*temp = milli_celsius;
> +
> +	return 0;
> +}
> +
> +static int adc_tm5_disable_channel(struct adc_tm5_channel *channel)
> +{
> +	struct adc_tm5_chip *chip = channel->chip;
> +	unsigned int reg = ADC_TM5_M_EN(channel->channel);
> +
> +	return adc_tm5_reg_update(chip, reg,
> +			ADC_TM5_M_MEAS_EN | ADC_TM5_M_HIGH_THR_INT_EN | ADC_TM5_M_LOW_THR_INT_EN,
> +			0);
> +}
> +
> +static int adc_tm5_configure(struct adc_tm5_channel *channel, int low_temp, int high_temp)
> +{
> +	struct adc_tm5_chip *chip = channel->chip;
> +	u8 buf[8];
> +	u16 reg = ADC_TM5_M_ADC_CH_SEL_CTL(channel->channel);
> +	int ret = 0;
> +
> +	ret = adc_tm5_read(chip, reg, buf, sizeof(buf));
> +	if (ret) {
> +		dev_err(chip->dev, "block read failed with %d\n", ret);
> +		return ret;
> +	}
> +
> +	/* Update ADC channel select */
> +	buf[0] = channel->adc_channel;
> +
> +	/* Warm temperature corresponds to low voltage threshold */
> +	if (high_temp != INT_MAX) {
> +		u16 adc_code = qcom_adc_tm5_temp_volt_scale(channel->prescale,
> +				chip->data->full_scale_code_volt, high_temp);
> +
> +		buf[1] = adc_code & 0xff;
> +		buf[2] = adc_code >> 8;
> +		buf[7] |= ADC_TM5_M_LOW_THR_INT_EN;
> +	}
> +
> +	/* Cool temperature corresponds to high voltage threshold */
> +	if (low_temp != -INT_MAX) {
> +		u16 adc_code = qcom_adc_tm5_temp_volt_scale(channel->prescale,
> +				chip->data->full_scale_code_volt, low_temp);
> +
> +		buf[3] = adc_code & 0xff;
> +		buf[4] = adc_code >> 8;
> +		buf[7] |= ADC_TM5_M_HIGH_THR_INT_EN;
> +	}
> +
> +	/* Update timer select */
> +	buf[5] = ADC5_TIMER_SEL_2;
> +
> +	/* Set calibration select, hw_settle delay */
> +	buf[6] &= ~ADC_TM5_M_CTL_HW_SETTLE_DELAY_MASK;
> +	buf[6] |= FIELD_PREP(ADC_TM5_M_CTL_HW_SETTLE_DELAY_MASK, channel->hw_settle_time);
> +	buf[6] &= ~ADC_TM5_M_CTL_CAL_SEL_MASK;
> +	buf[6] |= FIELD_PREP(ADC_TM5_M_CTL_CAL_SEL_MASK, channel->cal_method);
> +
> +	buf[7] |= ADC_TM5_M_MEAS_EN;
> +
> +	ret = adc_tm5_write(chip, reg, buf, sizeof(buf));
> +	if (ret)
> +		dev_err(chip->dev, "buf write failed\n");
> +
> +	return ret;
> +}
> +
> +static int adc_tm5_set_trips(void *data, int low_temp, int high_temp)
> +{
> +	struct adc_tm5_channel *channel = data;
> +	struct adc_tm5_chip *chip;
> +	int ret;
> +
> +	if (!channel)
> +		return -EINVAL;
> +
> +	chip = channel->chip;
> +	dev_dbg(chip->dev, "%d:low_temp(mdegC):%d, high_temp(mdegC):%d\n",
> +		channel->channel, low_temp, high_temp);
> +
> +	if (high_temp == INT_MAX && low_temp <= -INT_MAX)
> +		ret = adc_tm5_disable_channel(channel);
> +	else
> +		ret = adc_tm5_configure(channel, low_temp, high_temp);
In addition to the configurations done in adc_tm5_configure, you also 
need to write to the registers at 0x3546 (write 0x80 to enable the 
ADC_TM peripheral overall) and 0x3547 (this is the conversion request 
strobe, you need to write to bit 7 here too, to initiate the recurring 
measurements).
> +
> +	return ret;
> +}
> +



> +static int adc_tm5_probe(struct platform_device *pdev)
> +{
> +	struct device_node *node = pdev->dev.of_node;
> +	struct device *dev = &pdev->dev;
> +	struct adc_tm5_chip *adc_tm;
> +	struct regmap *regmap;
> +	int ret, irq;
> +	u32 reg;
> +
> +	regmap = dev_get_regmap(dev->parent, NULL);
> +	if (!regmap)
> +		return -ENODEV;
> +
> +	ret = of_property_read_u32(node, "reg", &reg);
> +	if (ret)
> +		return ret;
> +
> +	adc_tm = devm_kzalloc(&pdev->dev, sizeof(*adc_tm), GFP_KERNEL);
> +	if (!adc_tm)
> +		return -ENOMEM;
> +
> +	adc_tm->regmap = regmap;
> +	adc_tm->dev = dev;
> +	adc_tm->base = reg;
> +
> +	irq = platform_get_irq(pdev, 0);
> +	if (irq < 0) {
> +		dev_err(dev, "get_irq failed: %d\n", irq);
> +		return irq;
> +	}
> +
> +	ret = adc_tm5_get_dt_data(adc_tm, node);
> +	if (ret) {
> +		dev_err(dev, "get dt data failed: %d\n", ret);
> +		return ret;
> +	}
> +
> +	ret = adc_tm5_init(adc_tm);
> +	if (ret) {
> +		dev_err(dev, "adc-tm init failed\n");
> +		return ret;
> +	}
> +
> +	ret = adc_tm5_register_tzd(adc_tm);
> +	if (ret) {
> +		dev_err(dev, "tzd register failed\n");
> +		return ret;
> +	}
> +
> +	return devm_request_irq(dev, irq, adc_tm5_isr, 0, "pm-adc-tm5", adc_tm);
The interrupt should be requested with devm_request_threaded_irq, with 
the existing interrupt handler being given as the threaded function, for 
the reason mentioned above earlier.
> +}
> +
>

^ permalink raw reply	[flat|nested] 15+ messages in thread

end of thread, other threads:[~2020-09-25 11:42 UTC | newest]

Thread overview: 15+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-09-14 15:48 [PATCH v5 0/9] qcom: pm8150: add support for thermal monitoring Dmitry Baryshkov
2020-09-14 15:48 ` [PATCH v5 1/9] dt-bindings: thermal: qcom: add adc-thermal monitor bindings Dmitry Baryshkov
2020-09-22 23:40   ` Rob Herring
2020-09-23  9:07     ` Dmitry Baryshkov
2020-09-23 14:00       ` Rob Herring
2020-09-23 15:22         ` Jonathan Cameron
2020-09-14 15:48 ` [PATCH v5 2/9] fixp-arith: add a linear interpolation function Dmitry Baryshkov
2020-09-14 15:48 ` [PATCH v5 3/9] iio: adc: qcom-vadc: move several adc5 functions to common file Dmitry Baryshkov
2020-09-14 15:48 ` [PATCH v5 4/9] iio: adc: qcom-vadc-common: use fixp_linear_interpolate Dmitry Baryshkov
2020-09-14 15:48 ` [PATCH v5 5/9] iio: adc: qcom-spmi-adc5: use of_device_get_match_data Dmitry Baryshkov
2020-09-14 15:48 ` [PATCH v5 6/9] iio: provide of_iio_channel_get_by_name() and devm_ version it Dmitry Baryshkov
2020-09-14 15:48 ` [PATCH v5 7/9] thermal: qcom: add support for adc-tm5 PMIC thermal monitor Dmitry Baryshkov
2020-09-25 11:42   ` Jishnu Prakash
2020-09-14 15:48 ` [PATCH v5 8/9] arm64: dts: qcom: pm8150x: add definitions for adc-tm5 part Dmitry Baryshkov
2020-09-14 15:48 ` [PATCH v5 9/9] arm64: dts: sm8250-mtp: add thermal zones using pmic's adc-tm5 Dmitry Baryshkov

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).