[PATCH v5 0/2] iio: humidity: Add support for ENS210 sensor family

[PATCH v5 0/2] iio: humidity: Add support for ENS210 sensor family

[Joshua Felmeden]

This patch series adds support for the
ENS210/ENS210A/ENS211/ENS212/ENS213A/ENS215 temperature and humidity
sensors.

Patch 1 adds the required device tree bindings.

Patch 2 adds the driver, providing the probe and read functions.

Signed-off-by: Joshua Felmeden <jfelmeden@thegoodpenguin.co.uk>
---
changelog v4->v5:
- ens210.c: Updated ens210_data description to reflect values
- ens210.c: Updated dev_err to always use i2c device
- ens210.c: Check T_VALID before returning data
- ens210.c: Fix short linewrap on certain lines
- ens210.c: Added unreachable instead of unreachable return
- ens210.c: Fix sparse warnings
- ens210.c: Remove unneeded i2c_device_id
- ens210.c: Remove unneeded i2c_setclientdata()
- ens210.c: Simply return 'dev_err_probe()' instead of raw error code
- ens210.c: Fix dev_info print type

- Link to v4: https://lore.kernel.org/r/20240719-ens21x-v4-0-6044e48a376a@thegoodpenguin.co.uk
---
changelog v3->v4:
- rename ens21x.c -> ens210.c
- rename sciosense,ens21x.yaml -> sciosense,ens210.yaml
- General: Changed wildcard ens21x to ens210 to avoid wildcards
- Kconfig: added crc7 as a dependency
- Kconfig: remove extra blank lines
- ens210.c: Moved constants to inline
- ens210.c: Created ens210 chip info to handle chip specific conversion
- ens210.c: Added documentation for mutex
- ens210.c: Fixed sparse complaints from kernel bot
- ens210.c: Changed htonl to cpu_to_be32
- ens210.c: Renamed dev_data -> ens210_data
- ens210.c: changed of_match to i2c
- ens210.c: Added return check for i2c write
- ens210.c: Removed switch for conversion check, use chip_info instead
- ens210.c: Removed unnecessary retry loop
- ens210.c: Convert read to u8[3] type
- ens210.c: Ensure 'correct' path is inline, use early returns to break
    for errors
- ens210.c: Changed mutex to scoped_guard
- ens210.c: Changed part id mismatch to dev_info from dev_err and removed
    return
- ens210.c: Moved name from id->name to chip_info
- ens210.c: Changed data field to be chip_info rather than enum
- ens210.c: Formatting fixes (consistent spacing, extra line break at end
    removed)

- Link to v3: https://lore.kernel.org/r/20240710-ens21x-v3-0-4e3fbcf2a7fb@thegoodpenguin.co.uk
---
changelog v2->v3:
sciosense,ens21x.yaml: Update yaml to match dt_binding_check

- Link to V1: https://lore.kernel.org/all/20240709-ens21x-v1-2-678521433cdd@thegoodpenguin.co.uk/
- Link to v2: https://lore.kernel.org/r/20240710-ens21x-v2-0-a37c22018b0c@thegoodpenguin.co.uk

Many thanks for the feedback on the driver.
---
changelog v1->v2:
sciosense,ens21x.yaml: Add supply to documentation
sciosense,ens21x.yaml: Add fallback to compatible strings
ens21x.c: Move i2c_device_id next to of_device_id
ens21x.c: Use i2c_of_match_device() instead of of_match_device()

Many thanks for taking the time to review my patch.

Thanks,
Josh

To: Jonathan Cameron <jic23@kernel.org>
To: Lars-Peter Clausen <lars@metafoo.de>
To: Rob Herring <robh@kernel.org>
To: Krzysztof Kozlowski <krzk+dt@kernel.org>
To: Conor Dooley <conor+dt@kernel.org>
To: Joshua Felmeden <jfelmeden@thegoodpenguin.co.uk>
Cc: linux-iio@vger.kernel.org
Cc: devicetree@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
---

---
Joshua Felmeden (2):
      dt-bindings: iio: humidity: add ENS210 sensor family
      iio: humidity: Add support for ENS210

 .../bindings/iio/humidity/sciosense,ens210.yaml    |  55 ++++
 drivers/iio/humidity/Kconfig                       |  11 +
 drivers/iio/humidity/Makefile                      |   1 +
 drivers/iio/humidity/ens210.c                      | 344 +++++++++++++++++++++
 4 files changed, 411 insertions(+)
---
base-commit: 1ebab783647a9e3bf357002d5c4ff060c8474a0a
change-id: 20240709-ens21x-8f2530968f2e

Best regards,
-- 
Joshua Felmeden <jfelmeden@thegoodpenguin.co.uk>

[PATCH v5 1/2] dt-bindings: iio: humidity: add ENS210 sensor family

[Joshua Felmeden]

Add device tree documentation for ENS210 family of temperature and
humidity sensors

Signed-off-by: Joshua Felmeden <jfelmeden@thegoodpenguin.co.uk>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
---
 .../bindings/iio/humidity/sciosense,ens210.yaml    | 55 ++++++++++++++++++++++
 1 file changed, 55 insertions(+)

diff --git a/Documentation/devicetree/bindings/iio/humidity/sciosense,ens210.yaml b/Documentation/devicetree/bindings/iio/humidity/sciosense,ens210.yaml
new file mode 100644
index 000000000000..ed0ea938f7f8
--- /dev/null
+++ b/Documentation/devicetree/bindings/iio/humidity/sciosense,ens210.yaml
@@ -0,0 +1,55 @@
+# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/iio/humidity/sciosense,ens210.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: ScioSense ENS210 temperature and humidity sensor
+
+maintainers:
+  - Joshua Felmeden <jfelmeden@thegoodpenguin.co.uk>
+
+description: |
+  Temperature and Humidity sensor.
+
+  Datasheet:
+    https://www.sciosense.com/wp-content/uploads/2024/04/ENS21x-Datasheet.pdf
+    https://www.sciosense.com/wp-content/uploads/2023/12/ENS210-Datasheet.pdf
+
+properties:
+  compatible:
+    oneOf:
+      - items:
+          - enum:
+              - sciosense,ens210a
+              - sciosense,ens211
+              - sciosense,ens212
+              - sciosense,ens213a
+              - sciosense,ens215
+          - const: sciosense,ens210
+      - const: sciosense,ens210
+
+  reg:
+    maxItems: 1
+
+  vdd-supply: true
+
+required:
+  - compatible
+  - reg
+
+additionalProperties: false
+
+examples:
+  - |
+    i2c {
+       #address-cells = <1>;
+       #size-cells = <0>;
+
+       temperature-sensor@43 {
+           compatible = "sciosense,ens210";
+           reg = <0x43>;
+       };
+    };
+...
+

-- 
2.39.2

[PATCH v5 2/2] iio: humidity: Add support for ENS210

[Joshua Felmeden]

Add support for ENS210/ENS210A/ENS211/ENS212/ENS213A/ENS215.

The ENS21x is a family of temperature and relative humidity sensors with
accuracies tailored to the needs of specific applications.

Signed-off-by: Joshua Felmeden <jfelmeden@thegoodpenguin.co.uk>
---
 drivers/iio/humidity/Kconfig  |  11 ++
 drivers/iio/humidity/Makefile |   1 +
 drivers/iio/humidity/ens210.c | 344 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 356 insertions(+)

diff --git a/drivers/iio/humidity/Kconfig b/drivers/iio/humidity/Kconfig
index b15b7a3b66d5..54f11f000b6f 100644
--- a/drivers/iio/humidity/Kconfig
+++ b/drivers/iio/humidity/Kconfig
@@ -25,6 +25,17 @@ config DHT11
 	  Other sensors should work as well as long as they speak the
 	  same protocol.
 
+config ENS210
+	tristate "ENS210 temperature and humidity sensor"
+	depends on I2C
+	select CRC7
+	help
+	  Say yes here to get support for the ScioSense ENS210 family of
+	  humidity and temperature sensors.
+
+	  This driver can also be built as a module. If so, the module will be
+	  called ens210.
+
 config HDC100X
 	tristate "TI HDC100x relative humidity and temperature sensor"
 	depends on I2C
diff --git a/drivers/iio/humidity/Makefile b/drivers/iio/humidity/Makefile
index 5fbeef299f61..34b3dc749466 100644
--- a/drivers/iio/humidity/Makefile
+++ b/drivers/iio/humidity/Makefile
@@ -5,6 +5,7 @@
 
 obj-$(CONFIG_AM2315) += am2315.o
 obj-$(CONFIG_DHT11) += dht11.o
+obj-$(CONFIG_ENS210) += ens210.o
 obj-$(CONFIG_HDC100X) += hdc100x.o
 obj-$(CONFIG_HDC2010) += hdc2010.o
 obj-$(CONFIG_HDC3020) += hdc3020.o
diff --git a/drivers/iio/humidity/ens210.c b/drivers/iio/humidity/ens210.c
new file mode 100644
index 000000000000..9fe60a9eeef0
--- /dev/null
+++ b/drivers/iio/humidity/ens210.c
@@ -0,0 +1,344 @@
+// SPDX-License-Identifier: GPL-2.0+
+/*
+ * ens210.c - Support for ScioSense ens210 temperature & humidity sensor family
+ *
+ * (7-bit I2C slave address 0x43 ENS210)
+ * (7-bit I2C slave address 0x43 ENS210A)
+ * (7-bit I2C slave address 0x44 ENS211)
+ * (7-bit I2C slave address 0x45 ENS212)
+ * (7-bit I2C slave address 0x46 ENS213A)
+ * (7-bit I2C slave address 0x47 ENS215)
+ *
+ * Datasheet:
+ *  https://www.sciosense.com/wp-content/uploads/2024/04/ENS21x-Datasheet.pdf
+ *  https://www.sciosense.com/wp-content/uploads/2023/12/ENS210-Datasheet.pdf
+ */
+
+#include <linux/crc7.h>
+#include <linux/delay.h>
+#include <linux/i2c.h>
+#include <linux/iio/iio.h>
+#include <linux/mod_devicetable.h>
+#include <linux/module.h>
+#include <linux/types.h>
+
+#include <asm/unaligned.h>
+
+/* register definitions */
+#define ENS210_REG_PART_ID		0x00
+#define ENS210_REG_DIE_REV		0x02
+#define ENS210_REG_UID			0x04
+#define ENS210_REG_SYS_CTRL		0x10
+#define ENS210_REG_SYS_STAT		0x11
+#define ENS210_REG_SENS_RUN		0x21
+#define ENS210_REG_SENS_START		0x22
+#define ENS210_REG_SENS_STOP		0x23
+#define ENS210_REG_SENS_STAT		0x24
+#define ENS210_REG_T_VAL		0x30
+#define ENS210_REG_H_VAL		0x33
+
+/* value definitions */
+#define ENS210_SENS_START_T_START		BIT(0)
+#define ENS210_SENS_START_H_START		BIT(1)
+
+#define ENS210_SENS_STAT_T_ACTIVE		BIT(0)
+#define ENS210_SENS_STAT_H_ACTIVE		BIT(1)
+
+#define ENS210_SYS_CTRL_LOW_POWER_ENABLE	BIT(0)
+#define ENS210_SYS_CTRL_SYS_RESET		BIT(7)
+
+#define ENS210_SYS_STAT_SYS_ACTIVE		BIT(0)
+
+enum ens210_partnumber {
+	ENS210	= 0x0210,
+	ENS210A	= 0xa210,
+	ENS211	= 0x0211,
+	ENS212	= 0x0212,
+	ENS213A	= 0xa213,
+	ENS215	= 0x0215,
+};
+
+/**
+ * struct ens210_chip_info - Humidity/Temperature chip specific information
+ * @name:		name of device
+ * @part_id:		chip identifier
+ * @conv_time_msec:	time for conversion calculation in m/s
+ */
+struct ens210_chip_info {
+	const char *name;
+	enum ens210_partnumber part_id;
+	unsigned int conv_time_msec;
+};
+
+/**
+ * struct ens210_data - Humidity/Temperature sensor device structure
+ * @client:	i2c client
+ * @chip_info:	chip specific information
+ * @lock:	lock protecting the i2c conversion
+ */
+struct ens210_data {
+	struct i2c_client *client;
+	const struct ens210_chip_info *chip_info;
+	struct mutex lock;
+};
+
+/* calculate 17-bit crc7 */
+static u8 ens210_crc7(u32 val)
+{
+	unsigned int val_be = (val & 0x1ffff) >> 0x8;
+
+	return crc7_be(0xde, (u8 *)&val_be, 3) >> 1;
+}
+
+static int ens210_get_measurement(struct iio_dev *indio_dev, bool temp, int *val)
+{
+	struct ens210_data *data = iio_priv(indio_dev);
+	struct device *dev = &data->client->dev;
+	u32 regval;
+	u8 regval_le[3];
+	int ret;
+
+	/* assert read */
+	ret = i2c_smbus_write_byte_data(data->client, ENS210_REG_SENS_START,
+				  temp ? ENS210_SENS_START_T_START :
+					 ENS210_SENS_START_H_START);
+	if (ret)
+		return ret;
+
+	/* wait for conversion to be ready */
+	msleep(data->chip_info->conv_time_msec);
+
+	ret = i2c_smbus_read_byte_data(data->client,
+				       ENS210_REG_SENS_STAT);
+	if (ret < 0)
+		return ret;
+
+	/* perform read */
+	ret = i2c_smbus_read_i2c_block_data(
+		data->client, temp ? ENS210_REG_T_VAL : ENS210_REG_H_VAL, 3,
+		(u8 *)&regval_le);
+	if (ret < 0) {
+		dev_err(dev, "failed to read register");
+		return -EIO;
+	} else if (ret != 3) {
+		dev_err(dev, "expected 3 bytes, received %d\n", ret);
+		return -EIO;
+	}
+
+	regval = get_unaligned_le24(regval_le);
+	if (ens210_crc7(regval) != ((regval >> 17) & 0x7f)) {
+		/* crc fail */
+		dev_err(dev, "ens invalid crc\n");
+		return -EIO;
+	}
+
+	if (!((regval >> 16) & 0x1)) {
+		dev_err(dev, "ens data is not valid");
+		return -EIO;
+	}
+
+	*val = regval & 0xffff;
+	return IIO_VAL_INT;
+}
+
+static int ens210_read_raw(struct iio_dev *indio_dev,
+			   struct iio_chan_spec const *channel, int *val,
+			   int *val2, long mask)
+{
+	struct ens210_data *data = iio_priv(indio_dev);
+	int ret = -EINVAL;
+
+	switch (mask) {
+	case IIO_CHAN_INFO_RAW:
+		scoped_guard(mutex, &data->lock) {
+			ret = ens210_get_measurement(indio_dev,
+				channel->type == IIO_TEMP, val);
+			if (ret)
+				return ret;
+			return IIO_VAL_INT;
+		}
+		return ret;
+	case IIO_CHAN_INFO_SCALE:
+		if (channel->type == IIO_TEMP) {
+			*val = 15;
+			*val2 = 625000;
+		} else {
+			*val = 1;
+			*val2 = 953125;
+		}
+		return IIO_VAL_INT_PLUS_MICRO;
+	case IIO_CHAN_INFO_OFFSET:
+		if (channel->type == IIO_TEMP) {
+			*val = -17481;
+			*val2 = 600000;
+			ret = IIO_VAL_INT_PLUS_MICRO;
+			break;
+		}
+		*val = 0;
+		return IIO_VAL_INT;
+	default:
+		return -EINVAL;
+	}
+	unreachable();
+}
+
+static const struct iio_chan_spec ens210_channels[] = {
+	{
+		.type = IIO_TEMP,
+		.info_mask_separate = BIT(IIO_CHAN_INFO_RAW) |
+				      BIT(IIO_CHAN_INFO_SCALE) |
+				      BIT(IIO_CHAN_INFO_OFFSET),
+	},
+	{
+		.type = IIO_HUMIDITYRELATIVE,
+		.info_mask_separate = BIT(IIO_CHAN_INFO_RAW) |
+				      BIT(IIO_CHAN_INFO_SCALE) |
+				      BIT(IIO_CHAN_INFO_OFFSET),
+	}
+};
+
+static const struct iio_info ens210_info = {
+	.read_raw = ens210_read_raw,
+};
+
+static int ens210_probe(struct i2c_client *client)
+{
+	struct ens210_data *data;
+	struct iio_dev *indio_dev;
+	uint16_t part_id;
+	int ret;
+
+	if (!i2c_check_functionality(client->adapter,
+			I2C_FUNC_SMBUS_WRITE_BYTE_DATA |
+			I2C_FUNC_SMBUS_WRITE_BYTE |
+			I2C_FUNC_SMBUS_READ_I2C_BLOCK)) {
+		return dev_err_probe(&client->dev, -EOPNOTSUPP,
+			"adapter does not support some i2c transactions\n");
+	}
+
+	indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*data));
+	if (!indio_dev)
+		return -ENOMEM;
+
+	data = iio_priv(indio_dev);
+	data->client = client;
+	mutex_init(&data->lock);
+	data->chip_info = i2c_get_match_data(client);
+
+	ret = devm_regulator_get_enable(&client->dev, "vdd");
+	if (ret)
+		return ret;
+
+	/* reset device */
+	ret = i2c_smbus_write_byte_data(client, ENS210_REG_SYS_CTRL,
+					ENS210_SYS_CTRL_SYS_RESET);
+	if (ret)
+		return ret;
+
+	/* wait for device to become active */
+	usleep_range(4000, 5000);
+
+	/* disable low power mode */
+	ret = i2c_smbus_write_byte_data(client, ENS210_REG_SYS_CTRL, 0x00);
+	if (ret)
+		return ret;
+
+	/* wait for device to finish */
+	usleep_range(4000, 5000);
+
+	/* get part_id */
+	part_id = i2c_smbus_read_word_data(client, ENS210_REG_PART_ID);
+
+	if (part_id != data->chip_info->part_id) {
+		dev_info(&client->dev,
+			"Part ID does not match (0x%04x != 0x%04x)\n", part_id,
+			data->chip_info->part_id);
+	}
+
+	/* reenable low power */
+	ret = i2c_smbus_write_byte_data(client, ENS210_REG_SYS_CTRL,
+					ENS210_SYS_CTRL_LOW_POWER_ENABLE);
+	if (ret)
+		return ret;
+
+	indio_dev->name = data->chip_info->name;
+	indio_dev->modes = INDIO_DIRECT_MODE;
+	indio_dev->channels = ens210_channels;
+	indio_dev->num_channels = ARRAY_SIZE(ens210_channels);
+	indio_dev->info = &ens210_info;
+
+	return devm_iio_device_register(&client->dev, indio_dev);
+}
+
+static const struct ens210_chip_info ens210_chip_info_data = {
+	.name = "ens210",
+	.part_id = ENS210,
+	.conv_time_msec = 130,
+};
+
+static const struct ens210_chip_info ens210a_chip_info_data = {
+	.name = "ens210a",
+	.part_id = ENS210A,
+	.conv_time_msec = 130,
+};
+
+static const struct ens210_chip_info ens211_chip_info_data = {
+	.name = "ens211",
+	.part_id = ENS211,
+	.conv_time_msec = 32,
+};
+
+static const struct ens210_chip_info ens212_chip_info_data = {
+	.name = "ens212",
+	.part_id = ENS212,
+	.conv_time_msec = 32,
+};
+
+static const struct ens210_chip_info ens213a_chip_info_data = {
+	.name = "ens213a",
+	.part_id = ENS213A,
+	.conv_time_msec = 130,
+};
+
+static const struct ens210_chip_info ens215_chip_info_data = {
+	.name = "ens215",
+	.part_id = ENS215,
+	.conv_time_msec = 130,
+};
+
+static const struct of_device_id ens210_of_match[] = {
+	{ .compatible = "sciosense,ens210", .data = &ens210_chip_info_data},
+	{ .compatible = "sciosense,ens210a", .data = &ens210a_chip_info_data },
+	{ .compatible = "sciosense,ens211", .data = &ens211_chip_info_data},
+	{ .compatible = "sciosense,ens212", .data = &ens212_chip_info_data},
+	{ .compatible = "sciosense,ens213a", .data = &ens213a_chip_info_data },
+	{ .compatible = "sciosense,ens215", .data = &ens215_chip_info_data },
+	{}
+};
+MODULE_DEVICE_TABLE(of, ens210_of_match);
+
+static const struct i2c_device_id ens210_id_table[] = {
+	{ "ens210", (kernel_ulong_t)&ens210_chip_info_data },
+	{ "ens210a", (kernel_ulong_t)&ens210a_chip_info_data },
+	{ "ens211", (kernel_ulong_t)&ens211_chip_info_data },
+	{ "ens212", (kernel_ulong_t)&ens212_chip_info_data },
+	{ "ens213a", (kernel_ulong_t)&ens213a_chip_info_data },
+	{ "ens215", (kernel_ulong_t)&ens215_chip_info_data },
+	{}
+};
+MODULE_DEVICE_TABLE(i2c, ens210_id_table);
+
+static struct i2c_driver ens210_driver = {
+	.probe = ens210_probe,
+	.id_table = ens210_id_table,
+	.driver = {
+		.name = "ens210",
+		.of_match_table = ens210_of_match,
+	},
+};
+
+module_i2c_driver(ens210_driver);
+
+MODULE_DESCRIPTION("ScioSense ENS210 temperature and humidity sensor driver");
+MODULE_AUTHOR("Joshua Felmeden <jfelmeden@thegoodpenguin.co.uk>");
+MODULE_LICENSE("GPL");

-- 
2.39.2

Re: [PATCH v5 2/2] iio: humidity: Add support for ENS210

[Jonathan Cameron]

On Mon, 22 Jul 2024 17:58:43 +0100
Joshua Felmeden <jfelmeden@thegoodpenguin.co.uk> wrote:

> Add support for ENS210/ENS210A/ENS211/ENS212/ENS213A/ENS215.
> 
> The ENS21x is a family of temperature and relative humidity sensors with
> accuracies tailored to the needs of specific applications.
> 
> Signed-off-by: Joshua Felmeden <jfelmeden@thegoodpenguin.co.uk>
Hi Joshua,

I nearly took this with a few tweaks, but the address / casting issue for the
*_get_measurement looks to me like a more significant issue and the fix will need testing.

Various other minor bits and pieces inline.

Thanks,

Jonathan

> diff --git a/drivers/iio/humidity/ens210.c b/drivers/iio/humidity/ens210.c
> new file mode 100644
> index 000000000000..9fe60a9eeef0
> --- /dev/null
> +++ b/drivers/iio/humidity/ens210.c
> @@ -0,0 +1,344 @@

...

> +
> +/**
> + * struct ens210_data - Humidity/Temperature sensor device structure
> + * @client:	i2c client
> + * @chip_info:	chip specific information
> + * @lock:	lock protecting the i2c conversion

That's not enough detail. I2C has it's own locks so we don't need to protect the
bus. I assume this is really about protecting a particular sequence of multiple
accesses?

> + */
> +struct ens210_data {
> +	struct i2c_client *client;
> +	const struct ens210_chip_info *chip_info;
> +	struct mutex lock;
> +};
> +
> +/* calculate 17-bit crc7 */
> +static u8 ens210_crc7(u32 val)
> +{
> +	unsigned int val_be = (val & 0x1ffff) >> 0x8;
> +
> +	return crc7_be(0xde, (u8 *)&val_be, 3) >> 1;
> +}
> +
> +static int ens210_get_measurement(struct iio_dev *indio_dev, bool temp, int *val)
> +{
> +	struct ens210_data *data = iio_priv(indio_dev);
> +	struct device *dev = &data->client->dev;
> +	u32 regval;
> +	u8 regval_le[3];
> +	int ret;
> +
> +	/* assert read */
> +	ret = i2c_smbus_write_byte_data(data->client, ENS210_REG_SENS_START,
> +				  temp ? ENS210_SENS_START_T_START :
> +					 ENS210_SENS_START_H_START);
> +	if (ret)
> +		return ret;
> +
> +	/* wait for conversion to be ready */
> +	msleep(data->chip_info->conv_time_msec);
> +
> +	ret = i2c_smbus_read_byte_data(data->client,
> +				       ENS210_REG_SENS_STAT);

Under 80 chars on one line.

> +	if (ret < 0)
> +		return ret;
> +
> +	/* perform read */
> +	ret = i2c_smbus_read_i2c_block_data(
> +		data->client, temp ? ENS210_REG_T_VAL : ENS210_REG_H_VAL, 3,
> +		(u8 *)&regval_le);

This cast looks wrong and the need to have it indicates
a bigger issue.  regval_le is already u8 * so it's address is u8 ** and you
will be writing to the completely the wrong place.


> +	if (ret < 0) {
> +		dev_err(dev, "failed to read register");
> +		return -EIO;
you've returned so no need for an else for the next check.

	if (ret != 3)

> +	} else if (ret != 3) {
> +		dev_err(dev, "expected 3 bytes, received %d\n", ret);
> +		return -EIO;
> +	}
> +
> +	regval = get_unaligned_le24(regval_le);
> +	if (ens210_crc7(regval) != ((regval >> 17) & 0x7f)) {
> +		/* crc fail */
Comment very obvious. Drop it as such comments are just an opportunity for
becoming wrong in future as they bring no real value.

> +		dev_err(dev, "ens invalid crc\n");
> +		return -EIO;
> +	}
> +
> +	if (!((regval >> 16) & 0x1)) {
> +		dev_err(dev, "ens data is not valid");
> +		return -EIO;
> +	}
> +
> +	*val = regval & 0xffff;

	*val = regval & GENMASK(15, 0);
is, I think, slightly more readable.

> +	return IIO_VAL_INT;
> +}
> +
> +static int ens210_read_raw(struct iio_dev *indio_dev,
> +			   struct iio_chan_spec const *channel, int *val,
> +			   int *val2, long mask)
> +{
> +	struct ens210_data *data = iio_priv(indio_dev);
> +	int ret = -EINVAL;

If we hit this value something went wrong..  Indeed see missing unreachable
below.  If that is confusing the compiler add a comment to say so.

> +
> +	switch (mask) {
> +	case IIO_CHAN_INFO_RAW:
> +		scoped_guard(mutex, &data->lock) {
> +			ret = ens210_get_measurement(indio_dev,
> +				channel->type == IIO_TEMP, val);
> +			if (ret)
> +				return ret;
> +			return IIO_VAL_INT;
> +		}
> +		return ret;

unreachable()

> +	case IIO_CHAN_INFO_SCALE:
> +		if (channel->type == IIO_TEMP) {
> +			*val = 15;
> +			*val2 = 625000;
> +		} else {
> +			*val = 1;
> +			*val2 = 953125;
> +		}
> +		return IIO_VAL_INT_PLUS_MICRO;
> +	case IIO_CHAN_INFO_OFFSET:
> +		if (channel->type == IIO_TEMP) {
> +			*val = -17481;
> +			*val2 = 600000;
> +			ret = IIO_VAL_INT_PLUS_MICRO;
> +			break;

return IIO_VAL_INT_PLUS_MICRO;
not break to an unreachable()!

> +		}
> +		*val = 0;

The default offset is 0 so don't report one for the humidity channel.
Just drop bit being set in info_mask_separate below.


> +		return IIO_VAL_INT;
> +	default:
> +		return -EINVAL;
> +	}
> +	unreachable();
> +}
> +
> +static const struct iio_chan_spec ens210_channels[] = {
> +	{
> +		.type = IIO_TEMP,
> +		.info_mask_separate = BIT(IIO_CHAN_INFO_RAW) |
> +				      BIT(IIO_CHAN_INFO_SCALE) |
> +				      BIT(IIO_CHAN_INFO_OFFSET),
> +	},
> +	{
> +		.type = IIO_HUMIDITYRELATIVE,
> +		.info_mask_separate = BIT(IIO_CHAN_INFO_RAW) |
> +				      BIT(IIO_CHAN_INFO_SCALE) |
> +				      BIT(IIO_CHAN_INFO_OFFSET),
As above, drop the offset bit.
> +	}
> +};
> +
> +static const struct iio_info ens210_info = {
> +	.read_raw = ens210_read_raw,
> +};
> +
> +static int ens210_probe(struct i2c_client *client)
> +{
> +	struct ens210_data *data;
> +	struct iio_dev *indio_dev;
> +	uint16_t part_id;
> +	int ret;
> +
> +	if (!i2c_check_functionality(client->adapter,
> +			I2C_FUNC_SMBUS_WRITE_BYTE_DATA |
> +			I2C_FUNC_SMBUS_WRITE_BYTE |
> +			I2C_FUNC_SMBUS_READ_I2C_BLOCK)) {
> +		return dev_err_probe(&client->dev, -EOPNOTSUPP,
> +			"adapter does not support some i2c transactions\n");
> +	}
> +
> +	indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*data));
> +	if (!indio_dev)
> +		return -ENOMEM;
> +
> +	data = iio_priv(indio_dev);
> +	data->client = client;
> +	mutex_init(&data->lock);
> +	data->chip_info = i2c_get_match_data(client);
> +
> +	ret = devm_regulator_get_enable(&client->dev, "vdd");
> +	if (ret)
> +		return ret;
> +
> +	/* reset device */
> +	ret = i2c_smbus_write_byte_data(client, ENS210_REG_SYS_CTRL,
> +					ENS210_SYS_CTRL_SYS_RESET);
> +	if (ret)
> +		return ret;
> +
> +	/* wait for device to become active */
> +	usleep_range(4000, 5000);
> +
> +	/* disable low power mode */
> +	ret = i2c_smbus_write_byte_data(client, ENS210_REG_SYS_CTRL, 0x00);
> +	if (ret)
> +		return ret;
> +
> +	/* wait for device to finish */
> +	usleep_range(4000, 5000);
> +
> +	/* get part_id */
> +	part_id = i2c_smbus_read_word_data(client, ENS210_REG_PART_ID);
This may return an error. So should check negative value separately
as if that happens we don't want to carry on.
	ret = i2c_...
	if (ret < 0)
		return ret;
	part_id = ret;

	...


> +
> +	if (part_id != data->chip_info->part_id) {
> +		dev_info(&client->dev,
> +			"Part ID does not match (0x%04x != 0x%04x)\n", part_id,
> +			data->chip_info->part_id);
> +	}
> +
> +	/* reenable low power */
> +	ret = i2c_smbus_write_byte_data(client, ENS210_REG_SYS_CTRL,
> +					ENS210_SYS_CTRL_LOW_POWER_ENABLE);
> +	if (ret)
> +		return ret;
> +
> +	indio_dev->name = data->chip_info->name;
> +	indio_dev->modes = INDIO_DIRECT_MODE;
> +	indio_dev->channels = ens210_channels;
> +	indio_dev->num_channels = ARRAY_SIZE(ens210_channels);
> +	indio_dev->info = &ens210_info;
> +
> +	return devm_iio_device_register(&client->dev, indio_dev);
> +}

> +static const struct of_device_id ens210_of_match[] = {
> +	{ .compatible = "sciosense,ens210", .data = &ens210_chip_info_data},
> +	{ .compatible = "sciosense,ens210a", .data = &ens210a_chip_info_data },
> +	{ .compatible = "sciosense,ens211", .data = &ens211_chip_info_data},
> +	{ .compatible = "sciosense,ens212", .data = &ens212_chip_info_data},

Trivial: should always be a space before }

> +	{ .compatible = "sciosense,ens213a", .data = &ens213a_chip_info_data },
> +	{ .compatible = "sciosense,ens215", .data = &ens215_chip_info_data },
> +	{}
> +};
> +MODULE_DEVICE_TABLE(of, ens210_of_match);
> +
> +static const struct i2c_device_id ens210_id_table[] = {
> +	{ "ens210", (kernel_ulong_t)&ens210_chip_info_data },
> +	{ "ens210a", (kernel_ulong_t)&ens210a_chip_info_data },
> +	{ "ens211", (kernel_ulong_t)&ens211_chip_info_data },
> +	{ "ens212", (kernel_ulong_t)&ens212_chip_info_data },
> +	{ "ens213a", (kernel_ulong_t)&ens213a_chip_info_data },
> +	{ "ens215", (kernel_ulong_t)&ens215_chip_info_data },
> +	{}
> +};
> +MODULE_DEVICE_TABLE(i2c, ens210_id_table);
> +
> +static struct i2c_driver ens210_driver = {
> +	.probe = ens210_probe,
> +	.id_table = ens210_id_table,
> +	.driver = {
> +		.name = "ens210",
> +		.of_match_table = ens210_of_match,
> +	},
> +};
Trivial: Given close coupling between module_i2c_driver and the i2c_driver
structure it's common to not have a blank line here.

> +
> +module_i2c_driver(ens210_driver);
> +
> +MODULE_DESCRIPTION("ScioSense ENS210 temperature and humidity sensor driver");
> +MODULE_AUTHOR("Joshua Felmeden <jfelmeden@thegoodpenguin.co.uk>");
> +MODULE_LICENSE("GPL");
>