Skip to contentSkip to conversation

Add Ampere's Altra SMPro hwmon driver

PATCH Superseded Quan Nguyen 4 participants 12 messages Mar 29, 2021 – Apr 14, 2021

Subsystems (4)

v2 of 10 Superseded Latest is v10 →

lore
Share & tools
lore
#0 lore raw

v2 cover Add Ampere's Altra SMPro hwmon driver

This patch series adds support for Ampere SMpro hwmon driver. This driver supports accessing various CPU sensors provided by the SMpro co-processor including temperature, power, voltages, and current found on Ampere Altra processor family.

v2:
+ Used 'struct of_device_id's .data attribute [Lee Jones]
+ Removed "virtual" sensors [Guenter]
+ Fixed typo "mili" to "milli", "nanoWatt" to "microWatt" [Guenter]
+ Reported SOC_TDP as "Socket TDP" using max attributes [Guenter]
+ Clarified "highest" meaning in documentation [Guenter]
+ Corrected return error code when host is turn off [Guenter]
+ Reported MEM HOT Threshold for all DIMMs as temp*_crit [Guenter]
+ Removed license info as SPDX-License-Identifier existed [Guenter]
+ Added is_visible() support [Guenter]
+ Used HWMON_CHANNEL_INFO() macro and LABEL attributes [Guenter]
+ Made is_valid_id() return boolean [Guenter]
+ Returned -EPROBE_DEFER when smpro reg inaccessible [Guenter]
+ Removed unnecessary error message when dev register fail [Guenter]
+ Removed Socket TDP sensor [Quan]
+ Changed "ampere,ac01-smpro" to "ampere,smpro" [Quan]
+ Included sensor type and channel in labels [Quan]
+ Refactorized code to fix checkpatch.pl --strict complaint [Quan]

Quan Nguyen (4):
dt-bindings: mfd: Add bindings for Ampere Altra SMPro drivers
mfd: simple-mfd-i2c: Adds Ampere's Altra SMpro support
hwmon: smpro: Add Ampere's Altra smpro-hwmon driver
docs: hwmon: (smpro-hwmon) Add documentation

 .../bindings/hwmon/ampere,ac01-hwmon.yaml     |  27 +
 .../devicetree/bindings/mfd/ampere,smpro.yaml |  82 +++
 Documentation/hwmon/index.rst                 |   1 +
 Documentation/hwmon/smpro-hwmon.rst           | 101 ++++
 drivers/hwmon/Kconfig                         |   8 +
 drivers/hwmon/Makefile                        |   1 +
 drivers/hwmon/smpro-hwmon.c                   | 494 ++++++++++++++++++
 drivers/mfd/Kconfig                           |  10 +
 drivers/mfd/simple-mfd-i2c.c                  |   6 +
 9 files changed, 730 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/hwmon/ampere,ac01-hwmon.yaml
 create mode 100644 Documentation/devicetree/bindings/mfd/ampere,smpro.yaml
 create mode 100644 Documentation/hwmon/smpro-hwmon.rst
 create mode 100644 drivers/hwmon/smpro-hwmon.c
Signature · 1 line
2.28.0
#1 lore raw

PATCH v2 1/4 dt-bindings: mfd: Add bindings for Ampere Altra SMPro drivers

Adds device tree bindings for SMPro drivers found on the Mt.Jade hardware reference platform with Ampere's Altra Processor family.

---
 .../bindings/hwmon/ampere,ac01-hwmon.yaml     | 27 ++++++
 .../devicetree/bindings/mfd/ampere,smpro.yaml | 82 +++++++++++++++++++
 2 files changed, 109 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/hwmon/ampere,ac01-hwmon.yaml
 create mode 100644 Documentation/devicetree/bindings/mfd/ampere,smpro.yaml
Documentation/devicetree/bindings/hwmon/ampere,ac01-hwmon.yaml new file +27 -0
new file mode 100644
index 000000000000..015130a281f4
@@ -0,0 +1,27 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/hwmon/ampere,ac01-hwmon.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Hardware monitoring driver for the Ampere Altra SMPro
+
+maintainers:
+  - Quan Nguyen <[email protected]>
+
+description: |
+  This module is part of the Ampere Altra SMPro multi-function device. For more
+  details see ../mfd/ampere,smpro.yaml.
+
+properties:
+  compatible:
+    enum:
+      - ampere,ac01-hwmon
+
+  reg:
+    maxItems: 1
+
+required:
+  - compatible
+
+additionalProperties: false
Documentation/devicetree/bindings/mfd/ampere,smpro.yaml new file +82 -0

82 lines not on this page · raw message

Signature · 1 line
2.28.0
maintainer
#2 lore raw

Re: [PATCH v2 1/4] dt-bindings: mfd: Add bindings for Ampere Altra SMPro drivers

Quoted 58 lines · Quan Nguyen

Adds device tree bindings for SMPro drivers found on the Mt.Jade hardware reference platform with Ampere's Altra Processor family.

---
 .../bindings/hwmon/ampere,ac01-hwmon.yaml     | 27 ++++++
 .../devicetree/bindings/mfd/ampere,smpro.yaml | 82 +++++++++++++++++++
 2 files changed, 109 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/hwmon/ampere,ac01-hwmon.yaml
 create mode 100644 Documentation/devicetree/bindings/mfd/ampere,smpro.yaml
Documentation/devicetree/bindings/hwmon/ampere,ac01-hwmon.yaml new file +27 -0
new file mode 100644
index 000000000000..015130a281f4
@@ -0,0 +1,27 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/hwmon/ampere,ac01-hwmon.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Hardware monitoring driver for the Ampere Altra SMPro
+
+maintainers:
+  - Quan Nguyen <[email protected]>
+
+description: |
+  This module is part of the Ampere Altra SMPro multi-function device. For more
+  details see ../mfd/ampere,smpro.yaml.
+
+properties:
+  compatible:
+    enum:
+      - ampere,ac01-hwmon
+
+  reg:
+    maxItems: 1
+
+required:
+  - compatible
+
+additionalProperties: false
Documentation/devicetree/bindings/mfd/ampere,smpro.yaml new file +8 -0
new file mode 100644
index 000000000000..bf789c8a3d7d
@@ -0,0 +1,82 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/mfd/ampere,smpro.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Ampere Altra SMPro firmware driver
+

Documentation/devicetree/bindings/mfd/ampere,smpro.yaml line 18

Quoted 10 lines
Documentation/devicetree/bindings/mfd/ampere,smpro.yaml new file +10 -0
+maintainers:
+  - Quan Nguyen <[email protected]>
+
+description: |
+  Ampere Altra SMPro firmware may contain different blocks like hardware
+  monitoring, error monitoring and other miscellaneous features.
+
+properties:
+  compatible:
+    const: ampere,smpro

Only 1 version of SMPro? Needs to be more specific or provide details on how the exact version of firmware/hardware is discovered.

Quoted 9 lines
Documentation/devicetree/bindings/mfd/ampere,smpro.yaml new file +9 -0
+
+  reg:
+    description:
+      I2C device address.
+    maxItems: 1
+
+  "#address-cells":
+    const: 1
+

Documentation/devicetree/bindings/mfd/ampere,smpro.yaml line 37

Quoted 10 lines
Documentation/devicetree/bindings/mfd/ampere,smpro.yaml new file +10 -0
+  "#size-cells":
+    const: 0
+
+patternProperties:
+  "^hwmon(@[0-9a-f]+)?$":
+    $ref: ../hwmon/ampere,ac01-hwmon.yaml
+
+  "^misc(@[0-9a-f]+)?$":
+    type: object
+    description: Ampere Altra SMPro Misc driver

Bindings describe h/w, not drivers.

Quoted 32 lines
Documentation/devicetree/bindings/mfd/ampere,smpro.yaml new file +32 -0
+    properties:
+      compatible:
+        const: "ampere,ac01-misc"
+
+  "^errmon(@[0-9a-f]+)?$":
+    type: object
+    description: Ampere Altra SMPro Error Monitor driver
+    properties:
+      compatible:
+        const: "ampere,ac01-errmon"
+
+required:
+  - "#address-cells"
+  - "#size-cells"
+  - compatible
+  - reg
+
+additionalProperties: false
+
+examples:
+  - |
+    i2c {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        smpro at 4f {
+            compatible = "ampere,smpro";
+            reg = <0x4f>;
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            hwmon {

Documentation/devicetree/bindings/mfd/ampere,smpro.yaml line 79

Quoted 10 lines
Documentation/devicetree/bindings/mfd/ampere,smpro.yaml new file +10 -0
+                compatible = "ampere,ac01-hwmon";
+            };
+
+            misc {
+                compatible = "ampere,ac01-misc";
+            };
+
+            errmon {
+                compatible = "ampere,ac01-errmon";
+            };

None of the child nodes have any resources in DT, so you don't need them in DT.

Rob

#3 lore raw

Re: [PATCH v2 1/4] dt-bindings: mfd: Add bindings for Ampere Altra SMPro drivers

Hi Rob,
Thank you for reviewing

Quoted 64 lines · Rob Herring
Quoted 63 lines · Quan Nguyen

Adds device tree bindings for SMPro drivers found on the Mt.Jade hardware reference platform with Ampere's Altra Processor family.

---
  .../bindings/hwmon/ampere,ac01-hwmon.yaml     | 27 ++++++
  .../devicetree/bindings/mfd/ampere,smpro.yaml | 82 +++++++++++++++++++
  2 files changed, 109 insertions(+)

create mode 100644 Documentation/devicetree/bindings/hwmon/ampere,ac01-hwmon.yaml create mode 100644 Documentation/devicetree/bindings/mfd/ampere,smpro.yaml

Documentation/devicetree/bindings/hwmon/ampere,ac01-hwmon.yaml new file +27 -0
new file mode 100644
index 000000000000..015130a281f4
@@ -0,0 +1,27 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/hwmon/ampere,ac01-hwmon.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Hardware monitoring driver for the Ampere Altra SMPro
+
+maintainers:
+  - Quan Nguyen <[email protected]>
+
+description: |
+  This module is part of the Ampere Altra SMPro multi-function device. For more
+  details see ../mfd/ampere,smpro.yaml.
+
+properties:
+  compatible:
+    enum:
+      - ampere,ac01-hwmon
+
+  reg:
+    maxItems: 1
+
+required:
+  - compatible
+
+additionalProperties: false
Documentation/devicetree/bindings/mfd/ampere,smpro.yaml new file +13 -0
new file mode 100644
index 000000000000..bf789c8a3d7d
@@ -0,0 +1,82 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/mfd/ampere,smpro.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Ampere Altra SMPro firmware driver
+
+maintainers:
+  - Quan Nguyen <[email protected]>
+
+description: |
+  Ampere Altra SMPro firmware may contain different blocks like hardware
Quoted 9 lines

Documentation/devicetree/bindings/mfd/ampere,smpro.yaml line 18

Quoted 5 lines
Documentation/devicetree/bindings/mfd/ampere,smpro.yaml new file +5 -0
+  monitoring, error monitoring and other miscellaneous features.
+
+properties:
+  compatible:
+    const: ampere,smpro

Only 1 version of SMPro? Needs to be more specific or provide details on how the exact version of firmware/hardware is discovered.

Will change to enum in next version.

enum:
- ampere,smpro

Quoted 13 lines
Quoted 13 lines
Documentation/devicetree/bindings/mfd/ampere,smpro.yaml new file +13 -0
+
+  reg:
+    description:
+      I2C device address.
+    maxItems: 1
+
+  "#address-cells":
+    const: 1
+
+  "#size-cells":
+    const: 0
+
+patternProperties:
Quoted 9 lines

Documentation/devicetree/bindings/mfd/ampere,smpro.yaml line 37

Quoted 6 lines
Documentation/devicetree/bindings/mfd/ampere,smpro.yaml new file +6 -0
+  "^hwmon(@[0-9a-f]+)?$":
+    $ref: ../hwmon/ampere,ac01-hwmon.yaml
+
+  "^misc(@[0-9a-f]+)?$":
+    type: object
+    description: Ampere Altra SMPro Misc driver

Bindings describe h/w, not drivers.

Will fix in next version

Quoted 46 lines
Quoted 32 lines
Documentation/devicetree/bindings/mfd/ampere,smpro.yaml new file +32 -0
+    properties:
+      compatible:
+        const: "ampere,ac01-misc"
+
+  "^errmon(@[0-9a-f]+)?$":
+    type: object
+    description: Ampere Altra SMPro Error Monitor driver
+    properties:
+      compatible:
+        const: "ampere,ac01-errmon"
+
+required:
+  - "#address-cells"
+  - "#size-cells"
+  - compatible
+  - reg
+
+additionalProperties: false
+
+examples:
+  - |
+    i2c {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        smpro at 4f {
+            compatible = "ampere,smpro";
+            reg = <0x4f>;
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            hwmon {

Documentation/devicetree/bindings/mfd/ampere,smpro.yaml line 79

Quoted 10 lines
Documentation/devicetree/bindings/mfd/ampere,smpro.yaml new file +10 -0
+                compatible = "ampere,ac01-hwmon";
+            };
+
+            misc {
+                compatible = "ampere,ac01-misc";
+            };
+
+            errmon {
+                compatible = "ampere,ac01-errmon";
+            };

None of the child nodes have any resources in DT, so you don't need them in DT.

Quoted 2 lines

Rob

The intention was to re-use the drivers/mfd/simple-mfd-i2c.c and dont want to add extra codes to this driver just to instantiate these children. So for this case, the child drivers will not be instantiated if there are no child nodes in DT.

One solution I have in mind is to introduce resource (reg offset) for each child drivers.

-Quan

#4 lore raw

PATCH v2 2/4 mfd: simple-mfd-i2c: Adds Ampere's Altra SMpro support

In reply to Quan Nguyen

Adds an MFD driver for SMpro found on the Mt.Jade hardware reference platform with Ampere's Altra processor family.

---
 drivers/mfd/Kconfig          | 10 ++++++++++
 drivers/mfd/simple-mfd-i2c.c |  6 ++++++
 2 files changed, 16 insertions(+)
drivers/mfd/Kconfig +10 -0
index d07e8cf93286..f7a6460f7aa0 100644
@@ -77,6 +77,16 @@ config MFD_AS3711
 	help
 	  Support for the AS3711 PMIC from AMS
 
+config MFD_SMPRO
+	tristate "Ampere Computing MFD SMpro core driver"
+	select MFD_SIMPLE_MFD_I2C
+	help
+	  Say yes here to enable SMpro driver support for Ampere's Altra
+	  processor family.
+
+	  Ampere's Altra SMpro exposes an I2C regmap interface that can
+	  be accessed by child devices.
+
 config MFD_AS3722
 	tristate "ams AS3722 Power Management IC"
 	select MFD_CORE
drivers/mfd/simple-mfd-i2c.c +6 -0
index 87f684cff9a1..9a44655f5592 100644
@@ -21,6 +21,11 @@ static const struct regmap_config simple_regmap_config = {
 	.val_bits = 8,
 };
 
+static const struct regmap_config simple_word_regmap_config = {
+	.reg_bits = 8,
+	.val_bits = 16,
+};
+
 static int simple_mfd_i2c_probe(struct i2c_client *i2c)
 {
 	const struct regmap_config *config;
@@ -39,6 +44,7 @@ static int simple_mfd_i2c_probe(struct i2c_client *i2c)
 
 static const struct of_device_id simple_mfd_i2c_of_match[] = {
 	{ .compatible = "kontron,sl28cpld" },
+	{ .compatible = "ampere,smpro", .data = &simple_word_regmap_config },
 	{}
 };
 MODULE_DEVICE_TABLE(of, simple_mfd_i2c_of_match);
Signature · 1 line
2.28.0
#5 lore raw

Re: [PATCH v2 2/4] mfd: simple-mfd-i2c: Adds Ampere's Altra SMpro support

Quoted 8 lines · Quan Nguyen

Adds an MFD driver for SMpro found on the Mt.Jade hardware reference platform with Ampere's Altra processor family.

---
 drivers/mfd/Kconfig          | 10 ++++++++++
 drivers/mfd/simple-mfd-i2c.c |  6 ++++++
 2 files changed, 16 insertions(+)

For my own reference (apply this as-is to your sign-off block):

Signature · 4 lines
Lee Jones [???]
Senior Technical Lead - Developer Services
Linaro.org ? Open source software for Arm SoCs
Follow Linaro: Facebook | Twitter | Blog
#6 lore raw

PATCH v2 3/4 hwmon: smpro: Add Ampere's Altra smpro-hwmon driver

In reply to Quan Nguyen

This commit adds support for Ampere SMpro hwmon driver. This driver supports accessing various CPU sensors provided by the SMpro co-processor including temperature, power, voltages, and current.

---
 drivers/hwmon/Kconfig       |   8 +
 drivers/hwmon/Makefile      |   1 +
 drivers/hwmon/smpro-hwmon.c | 494 ++++++++++++++++++++++++++++++++++++
 3 files changed, 503 insertions(+)
 create mode 100644 drivers/hwmon/smpro-hwmon.c
drivers/hwmon/Kconfig +8 -0
index 0ddc974b102e..ba4b5a911baf 100644
@@ -67,6 +67,14 @@ config SENSORS_ABITUGURU3
 	  This driver can also be built as a module. If so, the module
 	  will be called abituguru3.
 
+config SENSORS_SMPRO
+	tristate "Ampere's Altra SMpro hardware monitoring driver"
+	depends on MFD_SMPRO
+	help
+	  If you say yes here you get support for the thermal, voltage,
+	  current and power sensors of Ampere's Altra processor family SoC
+	  with SMpro co-processor.
+
 config SENSORS_AD7314
 	tristate "Analog Devices AD7314 and compatibles"
 	depends on SPI
drivers/hwmon/Makefile +1 -0
index 59e78bc212cf..b25391f9c651 100644
@@ -174,6 +174,7 @@ obj-$(CONFIG_SENSORS_SHT3x)	+= sht3x.o
 obj-$(CONFIG_SENSORS_SHTC1)	+= shtc1.o
 obj-$(CONFIG_SENSORS_SIS5595)	+= sis5595.o
 obj-$(CONFIG_SENSORS_SMM665)	+= smm665.o
+obj-$(CONFIG_SENSORS_SMPRO)	+= smpro-hwmon.o
 obj-$(CONFIG_SENSORS_SMSC47B397)+= smsc47b397.o
 obj-$(CONFIG_SENSORS_SMSC47M1)	+= smsc47m1.o
 obj-$(CONFIG_SENSORS_SMSC47M192)+= smsc47m192.o
drivers/hwmon/smpro-hwmon.c new file +494 -0

494 lines not on this page · raw message

Signature · 1 line
2.28.0
maintainer
#7 lore raw

Re: [PATCH v2 3/4] hwmon: smpro: Add Ampere's Altra smpro-hwmon driver

Quoted 265 lines · Quan Nguyen

This commit adds support for Ampere SMpro hwmon driver. This driver supports accessing various CPU sensors provided by the SMpro co-processor including temperature, power, voltages, and current.

---
 drivers/hwmon/Kconfig       |   8 +
 drivers/hwmon/Makefile      |   1 +
 drivers/hwmon/smpro-hwmon.c | 494 ++++++++++++++++++++++++++++++++++++
 3 files changed, 503 insertions(+)
 create mode 100644 drivers/hwmon/smpro-hwmon.c
drivers/hwmon/Kconfig +8 -0
index 0ddc974b102e..ba4b5a911baf 100644
@@ -67,6 +67,14 @@ config SENSORS_ABITUGURU3
 	  This driver can also be built as a module. If so, the module
 	  will be called abituguru3.
 
+config SENSORS_SMPRO
+	tristate "Ampere's Altra SMpro hardware monitoring driver"
+	depends on MFD_SMPRO
+	help
+	  If you say yes here you get support for the thermal, voltage,
+	  current and power sensors of Ampere's Altra processor family SoC
+	  with SMpro co-processor.
+
 config SENSORS_AD7314
 	tristate "Analog Devices AD7314 and compatibles"
 	depends on SPI
drivers/hwmon/Makefile +1 -0
index 59e78bc212cf..b25391f9c651 100644
@@ -174,6 +174,7 @@ obj-$(CONFIG_SENSORS_SHT3x)	+= sht3x.o
 obj-$(CONFIG_SENSORS_SHTC1)	+= shtc1.o
 obj-$(CONFIG_SENSORS_SIS5595)	+= sis5595.o
 obj-$(CONFIG_SENSORS_SMM665)	+= smm665.o
+obj-$(CONFIG_SENSORS_SMPRO)	+= smpro-hwmon.o
 obj-$(CONFIG_SENSORS_SMSC47B397)+= smsc47b397.o
 obj-$(CONFIG_SENSORS_SMSC47M1)	+= smsc47m1.o
 obj-$(CONFIG_SENSORS_SMSC47M192)+= smsc47m192.o
drivers/hwmon/smpro-hwmon.c new file +216 -0
new file mode 100644
index 000000000000..4277736ebc6e
@@ -0,0 +1,494 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * Ampere Computing SoC's SMPro Hardware Monitoring Driver
+ *
+ * Copyright (c) 2021, Ampere Computing LLC
+ */
+#include <linux/bitfield.h>
+#include <linux/hwmon.h>
+#include <linux/hwmon-sysfs.h>
+#include <linux/kernel.h>
+#include <linux/mod_devicetable.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/property.h>
+#include <linux/regmap.h>
+
+/* Identification Registers */
+#define MANUFACTURER_ID_REG		0x02
+#define AMPERE_MANUFACTURER_ID		0xCD3A
+
+/* Logical Power Sensor Registers */
+#define SOC_TEMP_REG			0x10
+#define SOC_VRD_TEMP_REG		0x11
+#define DIMM_VRD_TEMP_REG		0x12
+#define CORE_VRD_TEMP_REG		0x13
+#define CH0_DIMM_TEMP_REG		0x14
+#define CH1_DIMM_TEMP_REG		0x15
+#define CH2_DIMM_TEMP_REG		0x16
+#define CH3_DIMM_TEMP_REG		0x17
+#define CH4_DIMM_TEMP_REG		0x18
+#define CH5_DIMM_TEMP_REG		0x19
+#define CH6_DIMM_TEMP_REG		0x1A
+#define CH7_DIMM_TEMP_REG		0x1B
+#define RCA_VRD_TEMP_REG		0x1C
+
+#define CORE_VRD_PWR_REG		0x20
+#define SOC_PWR_REG			0x21
+#define DIMM_VRD1_PWR_REG		0x22
+#define DIMM_VRD2_PWR_REG		0x23
+#define CORE_VRD_PWR_MW_REG		0x26
+#define SOC_PWR_MW_REG			0x27
+#define DIMM_VRD1_PWR_MW_REG		0x28
+#define DIMM_VRD2_PWR_MW_REG		0x29
+#define RCA_VRD_PWR_REG			0x2A
+#define RCA_VRD_PWR_MW_REG		0x2B
+
+#define MEM_HOT_THRESHOLD_REG		0x32
+#define SOC_VR_HOT_THRESHOLD_REG	0x33
+#define CORE_VRD_VOLT_REG		0x34
+#define SOC_VRD_VOLT_REG		0x35
+#define DIMM_VRD1_VOLT_REG		0x36
+#define DIMM_VRD2_VOLT_REG		0x37
+#define RCA_VRD_VOLT_REG		0x38
+
+#define CORE_VRD_CURR_REG		0x39
+#define SOC_VRD_CURR_REG		0x3A
+#define DIMM_VRD1_CURR_REG		0x3B
+#define DIMM_VRD2_CURR_REG		0x3C
+#define RCA_VRD_CURR_REG		0x3D
+
+struct smpro_hwmon {
+	struct regmap *regmap;
+};
+
+struct smpro_sensor {
+	const u8 reg;
+	const u8 reg_ext;
+	const char *label;
+};
+
+static const struct smpro_sensor temperature[] = {
+	{
+		.reg = SOC_TEMP_REG,
+		.label = "temp1 SoC"
+	},
+	{
+		.reg = SOC_VRD_TEMP_REG,
+		.label = "temp2 SoC VRD"
+	},
+	{
+		.reg = DIMM_VRD_TEMP_REG,
+		.label = "temp3 DIMM VRD"
+	},
+	{
+		.reg = CORE_VRD_TEMP_REG,
+		.label = "temp4 CORE VRD"
+	},
+	{
+		.reg = CH0_DIMM_TEMP_REG,
+		.label = "temp5 CH0 DIMM"
+	},
+	{
+		.reg = CH1_DIMM_TEMP_REG,
+		.label = "temp6 CH1 DIMM"
+	},
+	{
+		.reg = CH2_DIMM_TEMP_REG,
+		.label = "temp7 CH2 DIMM"
+	},
+	{
+		.reg = CH3_DIMM_TEMP_REG,
+		.label = "temp8 CH3 DIMM"
+	},
+	{
+		.reg = CH4_DIMM_TEMP_REG,
+		.label = "temp9 CH4 DIMM"
+	},
+	{
+		.reg = CH5_DIMM_TEMP_REG,
+		.label = "temp10 CH5 DIMM"
+	},
+	{
+		.reg = CH6_DIMM_TEMP_REG,
+		.label = "temp11 CH6 DIMM"
+	},
+	{
+		.reg = CH7_DIMM_TEMP_REG,
+		.label = "temp12 CH7 DIMM"
+	},
+	{
+		.reg = RCA_VRD_TEMP_REG,
+		.label = "temp13 RCA VRD"
+	},
+};
+
+static const struct smpro_sensor voltage[] = {
+	{
+		.reg = CORE_VRD_VOLT_REG,
+		.label = "vout0 CORE VRD"
+	},
+	{
+		.reg = SOC_VRD_VOLT_REG,
+		.label = "vout1 SoC VRD"
+	},
+	{
+		.reg = DIMM_VRD1_VOLT_REG,
+		.label = "vout2 DIMM VRD1"
+	},
+	{
+		.reg = DIMM_VRD2_VOLT_REG,
+		.label = "vout3 DIMM VRD2"
+	},
+	{
+		.reg = RCA_VRD_VOLT_REG,
+		.label = "vout4 RCA VRD"
+	},
+};
+
+static const struct smpro_sensor curr_sensor[] = {
+	{
+		.reg = CORE_VRD_CURR_REG,
+		.label = "iout1 CORE VRD"
+	},
+	{
+		.reg = SOC_VRD_CURR_REG,
+		.label = "iout2 SoC VRD"
+	},
+	{
+		.reg = DIMM_VRD1_CURR_REG,
+		.label = "iout3 DIMM VRD1"
+	},
+	{
+		.reg = DIMM_VRD2_CURR_REG,
+		.label = "iout4 DIMM VRD2"
+	},
+	{
+		.reg = RCA_VRD_CURR_REG,
+		.label = "iout5 RCA VRD"
+	},
+};
+
+static const struct smpro_sensor power[] = {
+	{
+		.reg = CORE_VRD_PWR_REG,
+		.reg_ext = CORE_VRD_PWR_MW_REG,
+		.label = "power1 CORE VRD"
+	},
+	{
+		.reg = SOC_PWR_REG,
+		.reg_ext = SOC_PWR_MW_REG,
+		.label = "power2 SoC"
+	},
+	{
+		.reg = DIMM_VRD1_PWR_REG,
+		.reg_ext = DIMM_VRD1_PWR_MW_REG,
+		.label = "power3 DIMM VRD1"
+	},
+	{
+		.reg = DIMM_VRD2_PWR_REG,
+		.reg_ext = DIMM_VRD2_PWR_MW_REG,
+		.label = "power4 DIMM VRD2"
+	},
+	{
+		.reg = RCA_VRD_PWR_REG,
+		.reg_ext = RCA_VRD_PWR_MW_REG,
+		.label = "power5 RCA VRD"
+	},
+};
+
+static int smpro_read_temp(struct device *dev, u32 attr, int channel, long *val)
+{
+	struct smpro_hwmon *hwmon = dev_get_drvdata(dev);
+	unsigned int value;
+	int ret;
+
+	switch (attr) {
+	case hwmon_temp_input:
+		ret = regmap_read(hwmon->regmap,
+				  temperature[channel].reg, &value);
+		if (ret)
+			return ret;
+		*val = (value & 0x1ff) * 1000;
+		break;
+	case hwmon_temp_crit:
+		if (temperature[channel].reg == SOC_VRD_TEMP_REG) {
+			ret = regmap_read(hwmon->regmap, SOC_VR_HOT_THRESHOLD_REG, &value);

drivers/hwmon/smpro-hwmon.c line 226

Quoted 10 lines
drivers/hwmon/smpro-hwmon.c new file +10 -0
+			if (ret)
+				return ret;
+			*val = (value & 0x1ff) * 1000;
+		} else {
+			/* Report same MEM HOT threshold across DIMM channels */
+			ret = regmap_read(hwmon->regmap, MEM_HOT_THRESHOLD_REG, &value);
+			if (ret)
+				return ret;
+			*val = (value & 0x1ff) * 1000;
+		}

To avoid code duplication:

		reg = temperature[channel].reg == SOC_VRD_TEMP_REG ? SOC_VR_HOT_THRESHOLD_REG : MEM_HOT_THRESHOLD_REG;
		ret = regmap_read(hwmon->regmap, reg, &value);
		if (ret)
			return ret;

But then why don't you just use reg_ext to store SOC_VR_HOT_THRESHOLD_REG or MEM_HOT_THRESHOLD_REG ? It is already available, after all, and with it the code could be simplified to

		ret = regmap_read(hwmon->regmap, temperature[channel].reg_ext, &value);
		if (ret)
			return ret;

I don't have a datasheet, but I do wonder what is in bit 9..15. Any idea ? Main question is if there is a sign bit, as theoretic as it may be.

Quoted 10 lines
drivers/hwmon/smpro-hwmon.c new file +10 -0
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+	return 0;
+}
+
+static int smpro_read_in(struct device *dev, u32 attr, int channel, long *val)
+{
+	struct smpro_hwmon *hwmon = dev_get_drvdata(dev);

drivers/hwmon/smpro-hwmon.c line 246

Quoted 10 lines
drivers/hwmon/smpro-hwmon.c new file +10 -0
+	unsigned int value;
+	int ret;
+
+	switch (attr) {
+	case hwmon_in_input:
+		ret = regmap_read(hwmon->regmap, voltage[channel].reg, &value);
+		if (ret < 0)
+			return ret;
+		/* Scale reported by the hardware is 1mV */
+		*val = value & 0x7fff;

What is in bit 15 ?

Quoted 61 lines
drivers/hwmon/smpro-hwmon.c new file +61 -0
+		return 0;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int smpro_read_curr(struct device *dev, u32 attr, int channel, long *val)
+{
+	struct smpro_hwmon *hwmon = dev_get_drvdata(dev);
+	unsigned int value;
+	int ret;
+
+	switch (attr) {
+	case hwmon_curr_input:
+		ret = regmap_read(hwmon->regmap, curr_sensor[channel].reg, &value);
+		if (ret < 0)
+			return ret;
+		/* Scale reported by the hardware is 1mA */
+		*val = value & 0x7fff;
+		return 0;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int smpro_read_power(struct device *dev, u32 attr, int channel, long *val_pwr)
+{
+	struct smpro_hwmon *hwmon = dev_get_drvdata(dev);
+	unsigned int val = 0, val_mw = 0;
+	int ret;
+
+	switch (attr) {
+	case hwmon_power_input:
+		ret = regmap_read(hwmon->regmap, power[channel].reg, &val);
+		if (ret)
+			return ret;
+
+		ret = regmap_read(hwmon->regmap, power[channel].reg_ext, &val_mw);
+		if (ret)
+			return ret;
+
+		*val_pwr = val * 1000000 + val_mw * 1000;
+		return 0;
+
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int smpro_read(struct device *dev, enum hwmon_sensor_types type,
+		      u32 attr, int channel, long *val)
+{
+	switch (type) {
+	case hwmon_temp:
+		return smpro_read_temp(dev, attr, channel, val);
+	case hwmon_in:
+		return smpro_read_in(dev, attr, channel, val);
+	case hwmon_power:
+		return smpro_read_power(dev, attr, channel, val);
+	case hwmon_curr:
+		return smpro_read_curr(dev, attr, channel, val);

drivers/hwmon/smpro-hwmon.c line 317

Quoted 10 lines
drivers/hwmon/smpro-hwmon.c new file +10 -0
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int smpro_write(struct device *dev, enum hwmon_sensor_types type,
+		       u32 attr, int channel, long val)
+{
+	return -EOPNOTSUPP;
+}

There are no writeable attributes, thus the write function is not needed.

Quoted 39 lines
drivers/hwmon/smpro-hwmon.c new file +39 -0
+
+static int smpro_read_string(struct device *dev, enum hwmon_sensor_types type,
+			     u32 attr, int channel, const char **str)
+{
+	switch (type) {
+	case hwmon_temp:
+		switch (attr) {
+		case hwmon_temp_label:
+			*str = temperature[channel].label;
+			return 0;
+		default:
+			return -EOPNOTSUPP;
+		}
+		break;
+
+	case hwmon_in:
+		switch (attr) {
+		case hwmon_in_label:
+			*str = voltage[channel].label;
+			return 0;
+		default:
+			return -EOPNOTSUPP;
+		}
+		break;
+
+	case hwmon_curr:
+		switch (attr) {
+		case hwmon_curr_label:
+			*str = curr_sensor[channel].label;
+			return 0;
+		default:
+			return -EOPNOTSUPP;
+		}
+		break;
+
+	case hwmon_power:
+		switch (attr) {
+		case hwmon_power_label:
+			*str = power[channel].label;

drivers/hwmon/smpro-hwmon.c line 366

Quoted 10 lines
drivers/hwmon/smpro-hwmon.c new file +10 -0
+			return 0;
+		default:
+			return -EOPNOTSUPP;
+		}
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return -EOPNOTSUPP;

If you are returning -ENOPSUPP by default, might as well replace all the same returns above with break;

Quoted 73 lines
drivers/hwmon/smpro-hwmon.c new file +73 -0
+}
+
+static umode_t smpro_is_visible(const void *data, enum hwmon_sensor_types type,
+				u32 attr, int channel)
+{
+	const struct smpro_hwmon *hwmon = data;
+	unsigned int value;
+	int ret;
+
+	switch (type) {
+	case hwmon_temp:
+		switch (attr) {
+		case hwmon_temp_input:
+		case hwmon_temp_label:
+		case hwmon_temp_crit:
+			ret = regmap_read(hwmon->regmap, temperature[channel].reg, &value);
+			if (ret || value == 0xFFFF)
+				return 0;
+		break;
+		}
+	default:
+		break;
+	}
+
+	return 0444;
+}
+
+static const struct hwmon_channel_info *smpro_info[] = {
+	HWMON_CHANNEL_INFO(temp,
+			   HWMON_T_INPUT | HWMON_T_LABEL,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL,
+			   HWMON_T_INPUT | HWMON_T_LABEL,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL),
+	HWMON_CHANNEL_INFO(in,
+			   HWMON_I_INPUT | HWMON_I_LABEL,
+			   HWMON_I_INPUT | HWMON_I_LABEL,
+			   HWMON_I_INPUT | HWMON_I_LABEL,
+			   HWMON_I_INPUT | HWMON_I_LABEL,
+			   HWMON_I_INPUT | HWMON_I_LABEL),
+	HWMON_CHANNEL_INFO(power,
+			   HWMON_P_INPUT | HWMON_P_LABEL,
+			   HWMON_P_INPUT | HWMON_P_LABEL,
+			   HWMON_P_INPUT | HWMON_P_LABEL,
+			   HWMON_P_INPUT | HWMON_P_LABEL,
+			   HWMON_P_INPUT | HWMON_P_LABEL),
+	HWMON_CHANNEL_INFO(curr,
+			   HWMON_C_INPUT | HWMON_C_LABEL,
+			   HWMON_C_INPUT | HWMON_C_LABEL,
+			   HWMON_C_INPUT | HWMON_C_LABEL,
+			   HWMON_C_INPUT | HWMON_C_LABEL,
+			   HWMON_C_INPUT | HWMON_C_LABEL),
+	NULL
+};
+
+static const struct hwmon_ops smpro_hwmon_ops = {
+	.is_visible = smpro_is_visible,
+	.read = smpro_read,
+	.write = smpro_write,
+	.read_string = smpro_read_string,
+};
+
+static const struct hwmon_chip_info smpro_chip_info = {
+	.ops = &smpro_hwmon_ops,
+	.info = smpro_info,

drivers/hwmon/smpro-hwmon.c line 449

Quoted 10 lines
drivers/hwmon/smpro-hwmon.c new file +10 -0
+};
+
+static bool is_valid_id(struct regmap *regmap)
+{
+	unsigned int val;
+	int ret;
+
+	ret = regmap_read(regmap, MANUFACTURER_ID_REG, &val);
+
+	return  (ret || (val != AMPERE_MANUFACTURER_ID)) ? false : true;

I am quite concerned about this: The calling code will translate it to -EPROBE_DEFER even if the manufacturer ID is wrong. It should return -ENODEV in that case. There should be a better means to determine if the controller is not available at all, or not yet.

Quoted 46 lines
drivers/hwmon/smpro-hwmon.c new file +45 -0
+}
+
+static int smpro_hwmon_probe(struct platform_device *pdev)
+{
+	struct smpro_hwmon *hwmon;
+	struct device *hwmon_dev;
+
+	hwmon = devm_kzalloc(&pdev->dev, sizeof(struct smpro_hwmon), GFP_KERNEL);
+	if (!hwmon)
+		return -ENOMEM;
+
+	hwmon->regmap = dev_get_regmap(pdev->dev.parent, NULL);
+	if (!hwmon->regmap)
+		return -ENODEV;
+
+	/* Check for valid ID */
+	if (!is_valid_id(hwmon->regmap))
+		return -EPROBE_DEFER;
+
+	hwmon_dev = devm_hwmon_device_register_with_info(&pdev->dev, "smpro_hwmon",
+							 hwmon, &smpro_chip_info, NULL);
+
+	return PTR_ERR_OR_ZERO(hwmon_dev);
+}
+
+static const struct of_device_id smpro_hwmon_of_match[] = {
+	{ .compatible = "ampere,ac01-hwmon" },
+	{}
+};
+MODULE_DEVICE_TABLE(of, smpro_hwmon_of_match);
+
+static struct platform_driver smpro_hwmon_driver = {
+	.probe		= smpro_hwmon_probe,
+	.driver = {
+		.name	= "smpro-hwmon",
+		.of_match_table = smpro_hwmon_of_match,
+	},
+};
+
+module_platform_driver(smpro_hwmon_driver);
+
+MODULE_AUTHOR("Thu Nguyen <[email protected]>");
+MODULE_AUTHOR("Quan Nguyen <[email protected]>");
+MODULE_DESCRIPTION("Ampere Altra SMPro hwmon driver");
+MODULE_LICENSE("GPL v2");
#8 lore raw

Re: [PATCH v2 3/4] hwmon: smpro: Add Ampere's Altra smpro-hwmon driver

Quoted 284 lines · Guenter Roeck
Quoted 265 lines · Quan Nguyen

This commit adds support for Ampere SMpro hwmon driver. This driver supports accessing various CPU sensors provided by the SMpro co-processor including temperature, power, voltages, and current.

---
  drivers/hwmon/Kconfig       |   8 +
  drivers/hwmon/Makefile      |   1 +
  drivers/hwmon/smpro-hwmon.c | 494 ++++++++++++++++++++++++++++++++++++
  3 files changed, 503 insertions(+)

create mode 100644 drivers/hwmon/smpro-hwmon.c

drivers/hwmon/Kconfig +8 -0
index 0ddc974b102e..ba4b5a911baf 100644
@@ -67,6 +67,14 @@ config SENSORS_ABITUGURU3
  	  This driver can also be built as a module. If so, the module
  	  will be called abituguru3.
 
+config SENSORS_SMPRO
+	tristate "Ampere's Altra SMpro hardware monitoring driver"
+	depends on MFD_SMPRO
+	help
+	  If you say yes here you get support for the thermal, voltage,
+	  current and power sensors of Ampere's Altra processor family SoC
+	  with SMpro co-processor.
+
  config SENSORS_AD7314
  	tristate "Analog Devices AD7314 and compatibles"
  	depends on SPI
drivers/hwmon/Makefile +1 -0
index 59e78bc212cf..b25391f9c651 100644
@@ -174,6 +174,7 @@ obj-$(CONFIG_SENSORS_SHT3x)	+= sht3x.o
  obj-$(CONFIG_SENSORS_SHTC1)	+= shtc1.o
  obj-$(CONFIG_SENSORS_SIS5595)	+= sis5595.o
  obj-$(CONFIG_SENSORS_SMM665)	+= smm665.o
+obj-$(CONFIG_SENSORS_SMPRO)	+= smpro-hwmon.o
  obj-$(CONFIG_SENSORS_SMSC47B397)+= smsc47b397.o
  obj-$(CONFIG_SENSORS_SMSC47M1)	+= smsc47m1.o
  obj-$(CONFIG_SENSORS_SMSC47M192)+= smsc47m192.o
drivers/hwmon/smpro-hwmon.c new file +216 -0
new file mode 100644
index 000000000000..4277736ebc6e
@@ -0,0 +1,494 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * Ampere Computing SoC's SMPro Hardware Monitoring Driver
+ *
+ * Copyright (c) 2021, Ampere Computing LLC
+ */
+#include <linux/bitfield.h>
+#include <linux/hwmon.h>
+#include <linux/hwmon-sysfs.h>
+#include <linux/kernel.h>
+#include <linux/mod_devicetable.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/property.h>
+#include <linux/regmap.h>
+
+/* Identification Registers */
+#define MANUFACTURER_ID_REG		0x02
+#define AMPERE_MANUFACTURER_ID		0xCD3A
+
+/* Logical Power Sensor Registers */
+#define SOC_TEMP_REG			0x10
+#define SOC_VRD_TEMP_REG		0x11
+#define DIMM_VRD_TEMP_REG		0x12
+#define CORE_VRD_TEMP_REG		0x13
+#define CH0_DIMM_TEMP_REG		0x14
+#define CH1_DIMM_TEMP_REG		0x15
+#define CH2_DIMM_TEMP_REG		0x16
+#define CH3_DIMM_TEMP_REG		0x17
+#define CH4_DIMM_TEMP_REG		0x18
+#define CH5_DIMM_TEMP_REG		0x19
+#define CH6_DIMM_TEMP_REG		0x1A
+#define CH7_DIMM_TEMP_REG		0x1B
+#define RCA_VRD_TEMP_REG		0x1C
+
+#define CORE_VRD_PWR_REG		0x20
+#define SOC_PWR_REG			0x21
+#define DIMM_VRD1_PWR_REG		0x22
+#define DIMM_VRD2_PWR_REG		0x23
+#define CORE_VRD_PWR_MW_REG		0x26
+#define SOC_PWR_MW_REG			0x27
+#define DIMM_VRD1_PWR_MW_REG		0x28
+#define DIMM_VRD2_PWR_MW_REG		0x29
+#define RCA_VRD_PWR_REG			0x2A
+#define RCA_VRD_PWR_MW_REG		0x2B
+
+#define MEM_HOT_THRESHOLD_REG		0x32
+#define SOC_VR_HOT_THRESHOLD_REG	0x33
+#define CORE_VRD_VOLT_REG		0x34
+#define SOC_VRD_VOLT_REG		0x35
+#define DIMM_VRD1_VOLT_REG		0x36
+#define DIMM_VRD2_VOLT_REG		0x37
+#define RCA_VRD_VOLT_REG		0x38
+
+#define CORE_VRD_CURR_REG		0x39
+#define SOC_VRD_CURR_REG		0x3A
+#define DIMM_VRD1_CURR_REG		0x3B
+#define DIMM_VRD2_CURR_REG		0x3C
+#define RCA_VRD_CURR_REG		0x3D
+
+struct smpro_hwmon {
+	struct regmap *regmap;
+};
+
+struct smpro_sensor {
+	const u8 reg;
+	const u8 reg_ext;
+	const char *label;
+};
+
+static const struct smpro_sensor temperature[] = {
+	{
+		.reg = SOC_TEMP_REG,
+		.label = "temp1 SoC"
+	},
+	{
+		.reg = SOC_VRD_TEMP_REG,
+		.label = "temp2 SoC VRD"
+	},
+	{
+		.reg = DIMM_VRD_TEMP_REG,
+		.label = "temp3 DIMM VRD"
+	},
+	{
+		.reg = CORE_VRD_TEMP_REG,
+		.label = "temp4 CORE VRD"
+	},
+	{
+		.reg = CH0_DIMM_TEMP_REG,
+		.label = "temp5 CH0 DIMM"
+	},
+	{
+		.reg = CH1_DIMM_TEMP_REG,
+		.label = "temp6 CH1 DIMM"
+	},
+	{
+		.reg = CH2_DIMM_TEMP_REG,
+		.label = "temp7 CH2 DIMM"
+	},
+	{
+		.reg = CH3_DIMM_TEMP_REG,
+		.label = "temp8 CH3 DIMM"
+	},
+	{
+		.reg = CH4_DIMM_TEMP_REG,
+		.label = "temp9 CH4 DIMM"
+	},
+	{
+		.reg = CH5_DIMM_TEMP_REG,
+		.label = "temp10 CH5 DIMM"
+	},
+	{
+		.reg = CH6_DIMM_TEMP_REG,
+		.label = "temp11 CH6 DIMM"
+	},
+	{
+		.reg = CH7_DIMM_TEMP_REG,
+		.label = "temp12 CH7 DIMM"
+	},
+	{
+		.reg = RCA_VRD_TEMP_REG,
+		.label = "temp13 RCA VRD"
+	},
+};
+
+static const struct smpro_sensor voltage[] = {
+	{
+		.reg = CORE_VRD_VOLT_REG,
+		.label = "vout0 CORE VRD"
+	},
+	{
+		.reg = SOC_VRD_VOLT_REG,
+		.label = "vout1 SoC VRD"
+	},
+	{
+		.reg = DIMM_VRD1_VOLT_REG,
+		.label = "vout2 DIMM VRD1"
+	},
+	{
+		.reg = DIMM_VRD2_VOLT_REG,
+		.label = "vout3 DIMM VRD2"
+	},
+	{
+		.reg = RCA_VRD_VOLT_REG,
+		.label = "vout4 RCA VRD"
+	},
+};
+
+static const struct smpro_sensor curr_sensor[] = {
+	{
+		.reg = CORE_VRD_CURR_REG,
+		.label = "iout1 CORE VRD"
+	},
+	{
+		.reg = SOC_VRD_CURR_REG,
+		.label = "iout2 SoC VRD"
+	},
+	{
+		.reg = DIMM_VRD1_CURR_REG,
+		.label = "iout3 DIMM VRD1"
+	},
+	{
+		.reg = DIMM_VRD2_CURR_REG,
+		.label = "iout4 DIMM VRD2"
+	},
+	{
+		.reg = RCA_VRD_CURR_REG,
+		.label = "iout5 RCA VRD"
+	},
+};
+
+static const struct smpro_sensor power[] = {
+	{
+		.reg = CORE_VRD_PWR_REG,
+		.reg_ext = CORE_VRD_PWR_MW_REG,
+		.label = "power1 CORE VRD"
+	},
+	{
+		.reg = SOC_PWR_REG,
+		.reg_ext = SOC_PWR_MW_REG,
+		.label = "power2 SoC"
+	},
+	{
+		.reg = DIMM_VRD1_PWR_REG,
+		.reg_ext = DIMM_VRD1_PWR_MW_REG,
+		.label = "power3 DIMM VRD1"
+	},
+	{
+		.reg = DIMM_VRD2_PWR_REG,
+		.reg_ext = DIMM_VRD2_PWR_MW_REG,
+		.label = "power4 DIMM VRD2"
+	},
+	{
+		.reg = RCA_VRD_PWR_REG,
+		.reg_ext = RCA_VRD_PWR_MW_REG,
+		.label = "power5 RCA VRD"
+	},
+};
+
+static int smpro_read_temp(struct device *dev, u32 attr, int channel, long *val)
+{
+	struct smpro_hwmon *hwmon = dev_get_drvdata(dev);
+	unsigned int value;
+	int ret;
+
+	switch (attr) {
+	case hwmon_temp_input:
+		ret = regmap_read(hwmon->regmap,
+				  temperature[channel].reg, &value);
+		if (ret)
+			return ret;
+		*val = (value & 0x1ff) * 1000;
+		break;
+	case hwmon_temp_crit:
+		if (temperature[channel].reg == SOC_VRD_TEMP_REG) {
+			ret = regmap_read(hwmon->regmap, SOC_VR_HOT_THRESHOLD_REG, &value);

drivers/hwmon/smpro-hwmon.c line 226

Quoted 10 lines
drivers/hwmon/smpro-hwmon.c new file +10 -0
+			if (ret)
+				return ret;
+			*val = (value & 0x1ff) * 1000;
+		} else {
+			/* Report same MEM HOT threshold across DIMM channels */
+			ret = regmap_read(hwmon->regmap, MEM_HOT_THRESHOLD_REG, &value);
+			if (ret)
+				return ret;
+			*val = (value & 0x1ff) * 1000;
+		}

To avoid code duplication:

		reg = temperature[channel].reg == SOC_VRD_TEMP_REG ? SOC_VR_HOT_THRESHOLD_REG : MEM_HOT_THRESHOLD_REG;
		ret = regmap_read(hwmon->regmap, reg, &value);
		if (ret)
			return ret;
Quoted 8 lines

But then why don't you just use reg_ext to store SOC_VR_HOT_THRESHOLD_REG or MEM_HOT_THRESHOLD_REG ? It is already available, after all, and with it the code could be simplified to

		ret = regmap_read(hwmon->regmap, temperature[channel].reg_ext, &value);
		if (ret)
			return ret;

Thank you for the comment.

Will change code follow this suggestion, will include in next version

Quoted 3 lines

I don't have a datasheet, but I do wonder what is in bit 9..15. Any idea ? Main question is if there is a sign bit, as theoretic as it may be.

The original intention was to use this as 9-bit 2-complement value follow LM75, but the fact is that the operation temperature is 0-125 C degree, so we simply use it as-is.

Quoted 14 lines
Quoted 14 lines
drivers/hwmon/smpro-hwmon.c new file +14 -0
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+	return 0;
+}
+
+static int smpro_read_in(struct device *dev, u32 attr, int channel, long *val)
+{
+	struct smpro_hwmon *hwmon = dev_get_drvdata(dev);
+	unsigned int value;
+	int ret;
+
+	switch (attr) {
Quoted 9 lines

drivers/hwmon/smpro-hwmon.c line 246

Quoted 6 lines
drivers/hwmon/smpro-hwmon.c new file +6 -0
+	case hwmon_in_input:
+		ret = regmap_read(hwmon->regmap, voltage[channel].reg, &value);
+		if (ret < 0)
+			return ret;
+		/* Scale reported by the hardware is 1mV */
+		*val = value & 0x7fff;

What is in bit 15 ?

This is 15-bit voltage in mV so the bit 15 (0-15) is unused.

Quoted 65 lines
Quoted 65 lines
drivers/hwmon/smpro-hwmon.c new file +65 -0
+		return 0;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int smpro_read_curr(struct device *dev, u32 attr, int channel, long *val)
+{
+	struct smpro_hwmon *hwmon = dev_get_drvdata(dev);
+	unsigned int value;
+	int ret;
+
+	switch (attr) {
+	case hwmon_curr_input:
+		ret = regmap_read(hwmon->regmap, curr_sensor[channel].reg, &value);
+		if (ret < 0)
+			return ret;
+		/* Scale reported by the hardware is 1mA */
+		*val = value & 0x7fff;
+		return 0;
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int smpro_read_power(struct device *dev, u32 attr, int channel, long *val_pwr)
+{
+	struct smpro_hwmon *hwmon = dev_get_drvdata(dev);
+	unsigned int val = 0, val_mw = 0;
+	int ret;
+
+	switch (attr) {
+	case hwmon_power_input:
+		ret = regmap_read(hwmon->regmap, power[channel].reg, &val);
+		if (ret)
+			return ret;
+
+		ret = regmap_read(hwmon->regmap, power[channel].reg_ext, &val_mw);
+		if (ret)
+			return ret;
+
+		*val_pwr = val * 1000000 + val_mw * 1000;
+		return 0;
+
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int smpro_read(struct device *dev, enum hwmon_sensor_types type,
+		      u32 attr, int channel, long *val)
+{
+	switch (type) {
+	case hwmon_temp:
+		return smpro_read_temp(dev, attr, channel, val);
+	case hwmon_in:
+		return smpro_read_in(dev, attr, channel, val);
+	case hwmon_power:
+		return smpro_read_power(dev, attr, channel, val);
+	case hwmon_curr:
+		return smpro_read_curr(dev, attr, channel, val);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
Quoted 9 lines

drivers/hwmon/smpro-hwmon.c line 317

Quoted 6 lines
drivers/hwmon/smpro-hwmon.c new file +6 -0
+
+static int smpro_write(struct device *dev, enum hwmon_sensor_types type,
+		       u32 attr, int channel, long val)
+{
+	return -EOPNOTSUPP;
+}

There are no writeable attributes, thus the write function is not needed.

Agree, will remove in next version

Quoted 44 lines
Quoted 44 lines
drivers/hwmon/smpro-hwmon.c new file +44 -0
+
+static int smpro_read_string(struct device *dev, enum hwmon_sensor_types type,
+			     u32 attr, int channel, const char **str)
+{
+	switch (type) {
+	case hwmon_temp:
+		switch (attr) {
+		case hwmon_temp_label:
+			*str = temperature[channel].label;
+			return 0;
+		default:
+			return -EOPNOTSUPP;
+		}
+		break;
+
+	case hwmon_in:
+		switch (attr) {
+		case hwmon_in_label:
+			*str = voltage[channel].label;
+			return 0;
+		default:
+			return -EOPNOTSUPP;
+		}
+		break;
+
+	case hwmon_curr:
+		switch (attr) {
+		case hwmon_curr_label:
+			*str = curr_sensor[channel].label;
+			return 0;
+		default:
+			return -EOPNOTSUPP;
+		}
+		break;
+
+	case hwmon_power:
+		switch (attr) {
+		case hwmon_power_label:
+			*str = power[channel].label;
+			return 0;
+		default:
+			return -EOPNOTSUPP;
+		}
+		break;
Quoted 9 lines

drivers/hwmon/smpro-hwmon.c line 249

Quoted 5 lines
drivers/hwmon/smpro-hwmon.c new file +5 -0
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return -EOPNOTSUPP;

If you are returning -ENOPSUPP by default, might as well replace all the same returns above with break;

Yes, will fix as you suggested. Will include in next version

Quoted 87 lines
Quoted 73 lines
drivers/hwmon/smpro-hwmon.c new file +73 -0
+}
+
+static umode_t smpro_is_visible(const void *data, enum hwmon_sensor_types type,
+				u32 attr, int channel)
+{
+	const struct smpro_hwmon *hwmon = data;
+	unsigned int value;
+	int ret;
+
+	switch (type) {
+	case hwmon_temp:
+		switch (attr) {
+		case hwmon_temp_input:
+		case hwmon_temp_label:
+		case hwmon_temp_crit:
+			ret = regmap_read(hwmon->regmap, temperature[channel].reg, &value);
+			if (ret || value == 0xFFFF)
+				return 0;
+		break;
+		}
+	default:
+		break;
+	}
+
+	return 0444;
+}
+
+static const struct hwmon_channel_info *smpro_info[] = {
+	HWMON_CHANNEL_INFO(temp,
+			   HWMON_T_INPUT | HWMON_T_LABEL,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL,
+			   HWMON_T_INPUT | HWMON_T_LABEL,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL | HWMON_T_CRIT,
+			   HWMON_T_INPUT | HWMON_T_LABEL),
+	HWMON_CHANNEL_INFO(in,
+			   HWMON_I_INPUT | HWMON_I_LABEL,
+			   HWMON_I_INPUT | HWMON_I_LABEL,
+			   HWMON_I_INPUT | HWMON_I_LABEL,
+			   HWMON_I_INPUT | HWMON_I_LABEL,
+			   HWMON_I_INPUT | HWMON_I_LABEL),
+	HWMON_CHANNEL_INFO(power,
+			   HWMON_P_INPUT | HWMON_P_LABEL,
+			   HWMON_P_INPUT | HWMON_P_LABEL,
+			   HWMON_P_INPUT | HWMON_P_LABEL,
+			   HWMON_P_INPUT | HWMON_P_LABEL,
+			   HWMON_P_INPUT | HWMON_P_LABEL),
+	HWMON_CHANNEL_INFO(curr,
+			   HWMON_C_INPUT | HWMON_C_LABEL,
+			   HWMON_C_INPUT | HWMON_C_LABEL,
+			   HWMON_C_INPUT | HWMON_C_LABEL,
+			   HWMON_C_INPUT | HWMON_C_LABEL,
+			   HWMON_C_INPUT | HWMON_C_LABEL),
+	NULL
+};
+
+static const struct hwmon_ops smpro_hwmon_ops = {
+	.is_visible = smpro_is_visible,
+	.read = smpro_read,
+	.write = smpro_write,
+	.read_string = smpro_read_string,
+};
+
+static const struct hwmon_chip_info smpro_chip_info = {
+	.ops = &smpro_hwmon_ops,
+	.info = smpro_info,

drivers/hwmon/smpro-hwmon.c line 449

Quoted 10 lines
drivers/hwmon/smpro-hwmon.c new file +10 -0
+};
+
+static bool is_valid_id(struct regmap *regmap)
+{
+	unsigned int val;
+	int ret;
+
+	ret = regmap_read(regmap, MANUFACTURER_ID_REG, &val);
+
+	return  (ret || (val != AMPERE_MANUFACTURER_ID)) ? false : true;

I am quite concerned about this: The calling code will translate it to

(quoted diff) +0 -2
-EPROBE_DEFER even if the manufacturer ID is wrong. It should return
-ENODEV in that case. There should be a better means to determine if the
Quoted 2 lines

controller is not available at all, or not yet.

Yes, I agree

Will fix in next version:
+ if the regmap_read return error, return -EPROBE_DEFER
+ if manufacturer ID is wrong, return -ENODEV

Quoted 47 lines
Quoted 46 lines
drivers/hwmon/smpro-hwmon.c new file +45 -0
+}
+
+static int smpro_hwmon_probe(struct platform_device *pdev)
+{
+	struct smpro_hwmon *hwmon;
+	struct device *hwmon_dev;
+
+	hwmon = devm_kzalloc(&pdev->dev, sizeof(struct smpro_hwmon), GFP_KERNEL);
+	if (!hwmon)
+		return -ENOMEM;
+
+	hwmon->regmap = dev_get_regmap(pdev->dev.parent, NULL);
+	if (!hwmon->regmap)
+		return -ENODEV;
+
+	/* Check for valid ID */
+	if (!is_valid_id(hwmon->regmap))
+		return -EPROBE_DEFER;
+
+	hwmon_dev = devm_hwmon_device_register_with_info(&pdev->dev, "smpro_hwmon",
+							 hwmon, &smpro_chip_info, NULL);
+
+	return PTR_ERR_OR_ZERO(hwmon_dev);
+}
+
+static const struct of_device_id smpro_hwmon_of_match[] = {
+	{ .compatible = "ampere,ac01-hwmon" },
+	{}
+};
+MODULE_DEVICE_TABLE(of, smpro_hwmon_of_match);
+
+static struct platform_driver smpro_hwmon_driver = {
+	.probe		= smpro_hwmon_probe,
+	.driver = {
+		.name	= "smpro-hwmon",
+		.of_match_table = smpro_hwmon_of_match,
+	},
+};
+
+module_platform_driver(smpro_hwmon_driver);
+
+MODULE_AUTHOR("Thu Nguyen <[email protected]>");
+MODULE_AUTHOR("Quan Nguyen <[email protected]>");
+MODULE_DESCRIPTION("Ampere Altra SMPro hwmon driver");
+MODULE_LICENSE("GPL v2");
maintainer
#9 lore raw

Re: [PATCH v2 3/4] hwmon: smpro: Add Ampere's Altra smpro-hwmon driver

On 4/7/21 12:41 AM, Quan Nguyen wrote:
[ ... ]

Quoted 9 lines
Quoted 9 lines

But then why don't you just use reg_ext to store SOC_VR_HOT_THRESHOLD_REG or MEM_HOT_THRESHOLD_REG ? It is already available, after all, and with it the code could be simplified to

??????? ret = regmap_read(hwmon->regmap, temperature[channel].reg_ext, &value); ??????? if (ret)
??????????? return ret;

Quoted 9 lines

Thank you for the comment.

Will change code follow this suggestion, will include in next version

Quoted 3 lines

I don't have a datasheet, but I do wonder what is in bit 9..15. Any idea ? Main question is if there is a sign bit, as theoretic as it may be.

The original intention was to use this as 9-bit 2-complement value follow LM75, but the fact is that the operation temperature is 0-125 C degree, so we simply use it as-is.

The operational temperature is not the question here. The question is if the chip _reports_ a sign. If it does, it should be handled, even if it is outside the operational range. The reported range is relevant here, not the operational range. After all, the chip won't really blow apart at -1 degrees C.

Thanks,
Guenter

#10 lore raw

Re: [PATCH v2 3/4] hwmon: smpro: Add Ampere's Altra smpro-hwmon driver

Quoted 18 lines · Guenter Roeck

On 4/7/21 12:41 AM, Quan Nguyen wrote:
[ ... ]

Quoted 16 lines
Quoted 9 lines

But then why don't you just use reg_ext to store SOC_VR_HOT_THRESHOLD_REG or MEM_HOT_THRESHOLD_REG ? It is already available, after all, and with it the code could be simplified to

??????? ret = regmap_read(hwmon->regmap, temperature[channel].reg_ext, &value); ??????? if (ret)
??????????? return ret;

Thank you for the comment.

Will change code follow this suggestion, will include in next version

Quoted 3 lines

I don't have a datasheet, but I do wonder what is in bit 9..15. Any idea ? Main question is if there is a sign bit, as theoretic as it may be.

Quoted 8 lines
Quoted 2 lines

The original intention was to use this as 9-bit 2-complement value follow LM75, but the fact is that the operation temperature is 0-125 C degree, so we simply use it as-is.

The operational temperature is not the question here. The question is if the chip _reports_ a sign. If it does, it should be handled, even if it is outside the operational range. The reported range is relevant here, not the operational range. After all, the chip won't really blow apart at -1 degrees C.

I think I've got it, will handle the sign bit in next version.

-Quan

#11 lore raw

PATCH v2 4/4 docs: hwmon: (smpro-hwmon) Add documentation

In reply to Quan Nguyen

Add documentation for the Ampere(R)'s Altra(R) SMpro hwmon driver.

---
 Documentation/hwmon/index.rst       |   1 +
 Documentation/hwmon/smpro-hwmon.rst | 101 ++++++++++++++++++++++++++++
 2 files changed, 102 insertions(+)
 create mode 100644 Documentation/hwmon/smpro-hwmon.rst
Documentation/hwmon/index.rst +1 -0
index 48bfa7887dd4..3e3631b253b6 100644
@@ -166,6 +166,7 @@ Hardware Monitoring Kernel Drivers
    sis5595
    sl28cpld
    smm665
+   smpro-hwmon
    smsc47b397
    smsc47m192
    smsc47m1
Documentation/hwmon/smpro-hwmon.rst new file +101 -0

101 lines not on this page · raw message

Signature · 1 line
2.28.0