[PATCH 0/2] Add driver for the TI TLA2528 i2c ADC

[PATCH 0/2] Add driver for the TI TLA2528 i2c ADC

[Maxime Chevallier]

Hi everyone,

This series is a resurection of a patch sent by Rodolfo Giometti that
adds support for the Texas Instruments TLA2528 8-channel 12bit ADC
accessed over i2c : [1]

With Rodolfo's permission, I've picked up the driver where it was left,
and made some changes :

 - All reviews that were made by Jonathan back in 2021 were addressed
 - I've updated the way we manage the regulator, using the newer
   devm_regulator_get_enable_read_voltage() helper
 - There was a bit of renaming for internal variables
 - I simplified a bit the probe/remove by using devm_iio_device_register
 - I've added a binding :)

This was tested on a custom board based on imx93, with a 3v3 ref supply.

Thanks Rodolfo for letting me give another shot at getting this driver
upstream, and to all the reviewers and maintainers that were involved in
the first round of reviews a few years ago.

Maxime

[1] : https://lore.kernel.org/linux-iio/20210802130441.972978-1-giometti@enneenne.com/

Maxime Chevallier (2):
  dt-bindings: iio: adc: add Texas Instruments TLA 2528
  iio: adc: add driver for Texas Instruments TLA2528 adc

 .../bindings/iio/adc/ti,tla2528.yaml          |  48 ++++
 MAINTAINERS                                   |   7 +
 drivers/iio/adc/Kconfig                       |  10 +
 drivers/iio/adc/Makefile                      |   1 +
 drivers/iio/adc/ti-tla2528.c                  | 209 ++++++++++++++++++
 5 files changed, 275 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml
 create mode 100644 drivers/iio/adc/ti-tla2528.c

-- 
2.49.0

[PATCH 1/2] dt-bindings: iio: adc: add Texas Instruments TLA 2528

[Maxime Chevallier]

The TI TLA 2528 is a simple 8 channel, 12-bit ADC? Add a binding
documentation for it.

Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
---
 .../bindings/iio/adc/ti,tla2528.yaml          | 48 +++++++++++++++++++
 1 file changed, 48 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml

diff --git a/Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml b/Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml
new file mode 100644
index 000000000000..0ee326d77014
--- /dev/null
+++ b/Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml
@@ -0,0 +1,48 @@
+# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/iio/adc/ti,tla2528.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Texas Instruments TLA2528 8-channel 12bit I2C ADC
+
+maintainers:
+  - Maxime Chevallier <maxime.chevallier@bootlin.com>
+
+description: |
+  12bit 8-channel I2C ADC.
+
+properties:
+  compatible:
+    const: ti,tla2528
+
+  reg:
+    maxItems: 1
+
+  vref-supply:
+    description: Supply for 2.35V to 5.5V reference voltage
+
+  "#io-channel-cells":
+    const: 1
+
+required:
+  - compatible
+  - reg
+  - vref-supply
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    i2c {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        adc@17 {
+            compatible = "ti,tla2528";
+            reg = <0x17>;
+            vref-supply = <&reg_adc_supply>;
+            #io-channel-cells = <1>;
+        };
+    };
+...
-- 
2.49.0

[PATCH 2/2] iio: adc: add driver for Texas Instruments TLA2528 adc

[Maxime Chevallier]

This adds a new driver for the TI TLA2528 ADC chip. It ha 8 12-bit
channels, that can also be configured as 16-bit averaging channels.

Add a very simple driver for it, allowing reading raw values for each
channel.

Signed-off-by: Rodolfo Giometti <giometti@enneenne.com>
Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
---
 MAINTAINERS                  |   7 ++
 drivers/iio/adc/Kconfig      |  10 ++
 drivers/iio/adc/Makefile     |   1 +
 drivers/iio/adc/ti-tla2528.c | 209 +++++++++++++++++++++++++++++++++++
 4 files changed, 227 insertions(+)
 create mode 100644 drivers/iio/adc/ti-tla2528.c

diff --git a/MAINTAINERS b/MAINTAINERS
index dc731d37c8fe..5c382ae216c7 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -25866,6 +25866,13 @@ F:	include/dt-bindings/soc/ti,sci_pm_domain.h
 F:	include/linux/soc/ti/ti_sci_inta_msi.h
 F:	include/linux/soc/ti/ti_sci_protocol.h
 
+TEXAS INSTRUMENTS' TLA2528 ADC DRIVER
+M:	Maxime Chevallier <maxime.chevallier@bootlin.com>
+L:	linux-iio@vger.kernel.org
+S:	Supported
+F:	Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml
+F:	drivers/iio/adc/ti-tla2528.c
+
 TEXAS INSTRUMENTS' TMP117 TEMPERATURE SENSOR DRIVER
 M:	Puranjay Mohan <puranjay@kernel.org>
 L:	linux-iio@vger.kernel.org
diff --git a/drivers/iio/adc/Kconfig b/drivers/iio/adc/Kconfig
index 58da8255525e..67376de410bf 100644
--- a/drivers/iio/adc/Kconfig
+++ b/drivers/iio/adc/Kconfig
@@ -1803,6 +1803,16 @@ config TI_LMP92064
 	  This driver can also be built as a module. If so, the module will be called
 	  ti-lmp92064.
 
+config TI_TLA2528
+	tristate "Texas Instruments TLA2528 ADC driver"
+	depends on I2C
+	help
+	  Say yes here to build support for Texas Instruments TLA2528
+	  12-Bit 8-Channel ADC.
+
+	  To compile this driver as a module, choose M here: the module will be
+	  called ti-tla2528.
+
 config TI_TLC4541
 	tristate "Texas Instruments TLC4541 ADC driver"
 	depends on SPI
diff --git a/drivers/iio/adc/Makefile b/drivers/iio/adc/Makefile
index 7cc8f9a12f76..941606defbf7 100644
--- a/drivers/iio/adc/Makefile
+++ b/drivers/iio/adc/Makefile
@@ -157,6 +157,7 @@ obj-$(CONFIG_TI_ADS8344) += ti-ads8344.o
 obj-$(CONFIG_TI_ADS8688) += ti-ads8688.o
 obj-$(CONFIG_TI_AM335X_ADC) += ti_am335x_adc.o
 obj-$(CONFIG_TI_LMP92064) += ti-lmp92064.o
+obj-$(CONFIG_TI_TLA2528) += ti-tla2528.o
 obj-$(CONFIG_TI_TLC4541) += ti-tlc4541.o
 obj-$(CONFIG_TI_TSC2046) += ti-tsc2046.o
 obj-$(CONFIG_TWL4030_MADC) += twl4030-madc.o
diff --git a/drivers/iio/adc/ti-tla2528.c b/drivers/iio/adc/ti-tla2528.c
new file mode 100644
index 000000000000..9c572e730ffb
--- /dev/null
+++ b/drivers/iio/adc/ti-tla2528.c
@@ -0,0 +1,209 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Driver for Texas Instruments TLA2528 ADC
+ *
+ * Copyright (C) 2020-2021 Rodolfo Giometti <giometti@enneenne.com>
+ * Copyright (C) 2025 Maxime Chevallier <maxime.chevallier@bootlin.com>
+ */
+
+#include <linux/delay.h>
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/mod_devicetable.h>
+#include <linux/regulator/consumer.h>
+
+#include <linux/iio/iio.h>
+
+#define TLA2528_OP_WRITE_REG		0x08
+
+#define TLA2528_DATA_CFG_ADR		0x02
+
+/* Datasheet says [5:4] sets the append status, but only bit 4 is used */
+#define TLA2528_DATA_CFG_APPEND_STATUS	BIT(4)
+#define TLA2528_PIN_CFG_ADR		0x05
+#define TLA2528_SEQUENCE_CFG_ADR	0x10
+#define TLA2528_CHANNEL_SEL_ADR		0x11
+
+struct tla2528 {
+	struct i2c_client *client;
+	int vref_uv;
+
+	/* Protects manual channel selection, i.e. last_read_channel */
+	struct mutex lock;
+	u8 last_read_channel;
+};
+
+static s32 tla2528_write_reg(const struct i2c_client *client, u8 reg, u8 val)
+{
+	u8 data[3] = {TLA2528_OP_WRITE_REG, reg, val};
+	int ret;
+
+	ret = i2c_master_send(client, data, 3);
+
+	return ret < 0 ? ret : 0;
+}
+
+static int tla2528_read_sample(const struct i2c_client *client)
+{
+	__be16 data;
+	int ret;
+
+	ret = i2c_master_recv(client, (char *)&data, 2);
+	if (ret < 0)
+		return ret;
+
+	return be16_to_cpu(data) >> 4;
+}
+
+static int tla2528_read(struct tla2528 *tla2528, u8 channel, int *val)
+{
+	struct i2c_client *client = tla2528->client;
+	int ret;
+
+	if (channel != tla2528->last_read_channel) {
+		ret = tla2528_write_reg(client, TLA2528_CHANNEL_SEL_ADR, channel);
+		if (ret < 0)
+			return ret;
+
+		tla2528->last_read_channel = channel;
+	}
+
+	ret = tla2528_read_sample(client);
+	if (ret < 0)
+		return ret;
+
+	*val = ret;
+
+	return 0;
+}
+
+static int tla2528_read_raw(struct iio_dev *indio_dev,
+			    struct iio_chan_spec const *chan,
+			    int *val, int *val2, long mask)
+{
+	struct tla2528 *tla2528 = iio_priv(indio_dev);
+	int ret;
+
+	switch (mask) {
+	case IIO_CHAN_INFO_RAW:
+		mutex_lock(&tla2528->lock);
+		ret = tla2528_read(tla2528, chan->channel, val);
+		mutex_unlock(&tla2528->lock);
+		if (ret < 0)
+			return ret;
+
+		return IIO_VAL_INT;
+
+	case IIO_CHAN_INFO_SCALE:
+		*val = tla2528->vref_uv / 1000;
+		*val2 = 12;
+
+		return IIO_VAL_FRACTIONAL_LOG2;
+
+	default:
+		return -EINVAL;
+	}
+}
+
+#define TLA2528_CHAN(_chan, _name) { \
+	.type = IIO_VOLTAGE,					\
+	.channel = (_chan),					\
+	.info_mask_separate = BIT(IIO_CHAN_INFO_RAW),		\
+	.info_mask_shared_by_type = BIT(IIO_CHAN_INFO_SCALE),	\
+	.datasheet_name = _name,				\
+	.indexed = 1,						\
+}
+
+static const struct iio_chan_spec tla2528_channel[] = {
+	TLA2528_CHAN(0, "AIN0"),
+	TLA2528_CHAN(1, "AIN1"),
+	TLA2528_CHAN(2, "AIN2"),
+	TLA2528_CHAN(3, "AIN3"),
+	TLA2528_CHAN(4, "AIN4"),
+	TLA2528_CHAN(5, "AIN5"),
+	TLA2528_CHAN(6, "AIN6"),
+	TLA2528_CHAN(7, "AIN7"),
+};
+
+static const struct iio_info tla2528_info = {
+	.read_raw = tla2528_read_raw,
+};
+
+static int tla2528_probe(struct i2c_client *client)
+{
+	struct iio_dev *indio_dev;
+	struct tla2528 *tla2528;
+	int ret;
+
+	if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C |
+				     I2C_FUNC_SMBUS_WRITE_WORD_DATA))
+		return -EOPNOTSUPP;
+
+	indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*tla2528));
+	if (!indio_dev)
+		return -ENOMEM;
+
+	tla2528 = iio_priv(indio_dev);
+	i2c_set_clientdata(client, indio_dev);
+	tla2528->client = client;
+
+	indio_dev->name = client->name;
+	indio_dev->info = &tla2528_info;
+	indio_dev->modes = INDIO_DIRECT_MODE;
+	indio_dev->channels = tla2528_channel;
+	indio_dev->num_channels = ARRAY_SIZE(tla2528_channel);
+
+	mutex_init(&tla2528->lock);
+
+	tla2528->vref_uv = devm_regulator_get_enable_read_voltage(&client->dev,
+								  "vref");
+	if (tla2528->vref_uv < 0)
+		return tla2528->vref_uv;
+
+	/* Set all inputs as analog */
+	ret = tla2528_write_reg(tla2528->client, TLA2528_PIN_CFG_ADR, 0x00);
+	if (ret < 0)
+		return ret;
+
+	ret = tla2528_write_reg(tla2528->client, TLA2528_DATA_CFG_ADR,
+				TLA2528_DATA_CFG_APPEND_STATUS);
+	if (ret < 0)
+		return ret;
+
+	/* Set manual mode */
+	ret = tla2528_write_reg(tla2528->client, TLA2528_SEQUENCE_CFG_ADR, 0x00);
+	if (ret < 0)
+		return ret;
+
+	/* Init private data */
+	tla2528->last_read_channel = ~0;
+
+	return devm_iio_device_register(&client->dev, indio_dev);
+}
+
+static const struct i2c_device_id tla2528_id[] = {
+	{ "tla2528", 0 },
+	{ }
+};
+MODULE_DEVICE_TABLE(i2c, tla2528_id);
+
+static const struct of_device_id tla2528_of_match[] = {
+	{ .compatible = "ti,tla2528", },
+	{  },
+};
+MODULE_DEVICE_TABLE(of, tla2528_of_match);
+
+static struct i2c_driver tla2528_driver = {
+	.driver = {
+		.name = "tla2528",
+		.of_match_table = tla2528_of_match,
+	},
+	.probe = tla2528_probe,
+	.id_table = tla2528_id,
+};
+module_i2c_driver(tla2528_driver);
+
+MODULE_AUTHOR("Maxime Chevallier <maxime.chevallier@bootlin.com>");
+MODULE_AUTHOR("Rodolfo Giometti <giometti@enneenne.com>");
+MODULE_DESCRIPTION("Texas Instruments TLA2528 ADC driver");
+MODULE_LICENSE("GPL");
-- 
2.49.0

Re: [PATCH 2/2] iio: adc: add driver for Texas Instruments TLA2528 adc

[Maxime Chevallier]

Hi again,

On 23/12/2025 16:55, Maxime Chevallier wrote:
> This adds a new driver for the TI TLA2528 ADC chip. It ha 8 12-bit
> channels, that can also be configured as 16-bit averaging channels.
> 
> Add a very simple driver for it, allowing reading raw values for each
> channel.
> 
> Signed-off-by: Rodolfo Giometti <giometti@enneenne.com>

Looking closer at this, Rodolfo hasn't seen this patch prior to me
sending it, so it should rather be :

Orginally-by: Rodolfo Giometti <giometti@enneenne.com>

> Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>

I can resend if need be :)

Maxime

Re: [PATCH 2/2] iio: adc: add driver for Texas Instruments TLA2528 adc

[David Lechner]

On 12/23/25 11:12 AM, Maxime Chevallier wrote:
> Hi again,
> 
> On 23/12/2025 16:55, Maxime Chevallier wrote:
>> This adds a new driver for the TI TLA2528 ADC chip. It ha 8 12-bit
>> channels, that can also be configured as 16-bit averaging channels.
>>
>> Add a very simple driver for it, allowing reading raw values for each
>> channel.
>>
>> Signed-off-by: Rodolfo Giometti <giometti@enneenne.com>
> 
> Looking closer at this, Rodolfo hasn't seen this patch prior to me
> sending it, so it should rather be :
> 
> Orginally-by: Rodolfo Giometti <giometti@enneenne.com>

I think the usual way would be to keep the Signed-off-by: and add
Co-developed-by:.

See https://docs.kernel.org/process/submitting-patches.html#when-to-use-acked-by-cc-and-co-developed-by

> 
>> Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
> 
> I can resend if need be :)

Wait a week for other reviews. :-)

> 
> Maxime
> 

Re: [PATCH 1/2] dt-bindings: iio: adc: add Texas Instruments TLA 2528

[David Lechner]

On 12/23/25 9:55 AM, Maxime Chevallier wrote:
> The TI TLA 2528 is a simple 8 channel, 12-bit ADC? Add a binding

TLA2528 (no space). Also, why the "?"?

> documentation for it.
> 
> Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
> ---
>  .../bindings/iio/adc/ti,tla2528.yaml          | 48 +++++++++++++++++++
>  1 file changed, 48 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml
> 
> diff --git a/Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml b/Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml
> new file mode 100644
> index 000000000000..0ee326d77014
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml
> @@ -0,0 +1,48 @@
> +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
> +%YAML 1.2
> +---
> +$id: http://devicetree.org/schemas/iio/adc/ti,tla2528.yaml#
> +$schema: http://devicetree.org/meta-schemas/core.yaml#
> +
> +title: Texas Instruments TLA2528 8-channel 12bit I2C ADC

12-bit

> +
> +maintainers:
> +  - Maxime Chevallier <maxime.chevallier@bootlin.com>
> +
> +description: |
> +  12bit 8-channel I2C ADC.

The title already says this. Either drop it or add new info.

Also, don't need the |.

> +
> +properties:
> +  compatible:
> +    const: ti,tla2528
> +
> +  reg:
> +    maxItems: 1
> +
> +  vref-supply:
> +    description: Supply for 2.35V to 5.5V reference voltage

According the the datasheet, there are AVDD and DVDD supplies.
Nothing named VREF or REF.

So instead:

avdd-supply: true
dvdd-supply: true


It looks like inputs can also be used as GPIOs, so

gpio-controller: true
#gpio-cells:
  const: 2

would be appropriate (it doesn't matter if the driver doesn't
implement it, we know what the correct bindings are).

> +
> +  "#io-channel-cells":
> +    const: 1
> +
> +required:
> +  - compatible
> +  - reg
> +  - vref-supply
> +
> +unevaluatedProperties: false
> +
> +examples:
> +  - |
> +    i2c {
> +        #address-cells = <1>;
> +        #size-cells = <0>;
> +
> +        adc@17 {
> +            compatible = "ti,tla2528";
> +            reg = <0x17>;
> +            vref-supply = <&reg_adc_supply>;
> +            #io-channel-cells = <1>;
> +        };
> +    };
> +...

Re: [PATCH 2/2] iio: adc: add driver for Texas Instruments TLA2528 adc

[David Lechner]

On 12/23/25 9:55 AM, Maxime Chevallier wrote:
> This adds a new driver for the TI TLA2528 ADC chip. It ha 8 12-bit

s/This adds/Add/

s/ha/has/

> channels, that can also be configured as 16-bit averaging channels.
> 
> Add a very simple driver for it, allowing reading raw values for each

Don't need to say we are adding a driver twice.

> channel.
> 
> Signed-off-by: Rodolfo Giometti <giometti@enneenne.com>
> Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
> ---
>  MAINTAINERS                  |   7 ++
>  drivers/iio/adc/Kconfig      |  10 ++
>  drivers/iio/adc/Makefile     |   1 +
>  drivers/iio/adc/ti-tla2528.c | 209 +++++++++++++++++++++++++++++++++++
>  4 files changed, 227 insertions(+)
>  create mode 100644 drivers/iio/adc/ti-tla2528.c
> 
> diff --git a/MAINTAINERS b/MAINTAINERS
> index dc731d37c8fe..5c382ae216c7 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -25866,6 +25866,13 @@ F:	include/dt-bindings/soc/ti,sci_pm_domain.h
>  F:	include/linux/soc/ti/ti_sci_inta_msi.h
>  F:	include/linux/soc/ti/ti_sci_protocol.h
>  
> +TEXAS INSTRUMENTS' TLA2528 ADC DRIVER
> +M:	Maxime Chevallier <maxime.chevallier@bootlin.com>
> +L:	linux-iio@vger.kernel.org
> +S:	Supported
> +F:	Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml

This part can be included in the dt-bindings patch since that is where
the file is introduced.

> +F:	drivers/iio/adc/ti-tla2528.c

And just keep this line in this patch.

> +
>  TEXAS INSTRUMENTS' TMP117 TEMPERATURE SENSOR DRIVER
>  M:	Puranjay Mohan <puranjay@kernel.org>
>  L:	linux-iio@vger.kernel.org
> diff --git a/drivers/iio/adc/Kconfig b/drivers/iio/adc/Kconfig
> index 58da8255525e..67376de410bf 100644
> --- a/drivers/iio/adc/Kconfig
> +++ b/drivers/iio/adc/Kconfig
> @@ -1803,6 +1803,16 @@ config TI_LMP92064
>  	  This driver can also be built as a module. If so, the module will be called
>  	  ti-lmp92064.
>  
> +config TI_TLA2528
> +	tristate "Texas Instruments TLA2528 ADC driver"
> +	depends on I2C
> +	help
> +	  Say yes here to build support for Texas Instruments TLA2528
> +	  12-Bit 8-Channel ADC.
> +
> +	  To compile this driver as a module, choose M here: the module will be
> +	  called ti-tla2528.
> +
>  config TI_TLC4541
>  	tristate "Texas Instruments TLC4541 ADC driver"
>  	depends on SPI
> diff --git a/drivers/iio/adc/Makefile b/drivers/iio/adc/Makefile
> index 7cc8f9a12f76..941606defbf7 100644
> --- a/drivers/iio/adc/Makefile
> +++ b/drivers/iio/adc/Makefile
> @@ -157,6 +157,7 @@ obj-$(CONFIG_TI_ADS8344) += ti-ads8344.o
>  obj-$(CONFIG_TI_ADS8688) += ti-ads8688.o
>  obj-$(CONFIG_TI_AM335X_ADC) += ti_am335x_adc.o
>  obj-$(CONFIG_TI_LMP92064) += ti-lmp92064.o
> +obj-$(CONFIG_TI_TLA2528) += ti-tla2528.o
>  obj-$(CONFIG_TI_TLC4541) += ti-tlc4541.o
>  obj-$(CONFIG_TI_TSC2046) += ti-tsc2046.o
>  obj-$(CONFIG_TWL4030_MADC) += twl4030-madc.o
> diff --git a/drivers/iio/adc/ti-tla2528.c b/drivers/iio/adc/ti-tla2528.c
> new file mode 100644
> index 000000000000..9c572e730ffb
> --- /dev/null
> +++ b/drivers/iio/adc/ti-tla2528.c
> @@ -0,0 +1,209 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Driver for Texas Instruments TLA2528 ADC
> + *
> + * Copyright (C) 2020-2021 Rodolfo Giometti <giometti@enneenne.com>
> + * Copyright (C) 2025 Maxime Chevallier <maxime.chevallier@bootlin.com>
> + */
> +
> +#include <linux/delay.h>

Check the headers to see what is actually used. For example,
I don't seen any delays/sleeps. And we have a mutex, but no
mutex header.

> +#include <linux/i2c.h>
> +#include <linux/module.h>
> +#include <linux/mod_devicetable.h>
> +#include <linux/regulator/consumer.h>
> +
> +#include <linux/iio/iio.h>
> +
> +#define TLA2528_OP_WRITE_REG		0x08
> +
> +#define TLA2528_DATA_CFG_ADR		0x02
> +
> +/* Datasheet says [5:4] sets the append status, but only bit 4 is used */
> +#define TLA2528_DATA_CFG_APPEND_STATUS	BIT(4)
> +#define TLA2528_PIN_CFG_ADR		0x05
> +#define TLA2528_SEQUENCE_CFG_ADR	0x10
> +#define TLA2528_CHANNEL_SEL_ADR		0x11
> +
> +struct tla2528 {
> +	struct i2c_client *client;
> +	int vref_uv;
> +
> +	/* Protects manual channel selection, i.e. last_read_channel */
> +	struct mutex lock;
> +	u8 last_read_channel;
> +};
> +
> +static s32 tla2528_write_reg(const struct i2c_client *client, u8 reg, u8 val)

Usually type is just "int" for error code returns.

> +{
> +	u8 data[3] = {TLA2528_OP_WRITE_REG, reg, val};
> +	int ret;
> +
> +	ret = i2c_master_send(client, data, 3);
> +
> +	return ret < 0 ? ret : 0;
> +}
> +

Would it make sense to use regmap instead?


> +static int tla2528_read_sample(const struct i2c_client *client)
> +{
> +	__be16 data;
> +	int ret;
> +

Don't we need to set the CNVST bit in GENERAL_CFG register to trigger
a conversion in manual mode?

> +	ret = i2c_master_recv(client, (char *)&data, 2);
> +	if (ret < 0)
> +		return ret;
> +
> +	return be16_to_cpu(data) >> 4;
> +}
> +
> +static int tla2528_read(struct tla2528 *tla2528, u8 channel, int *val)
> +{
> +	struct i2c_client *client = tla2528->client;
> +	int ret;
> +
> +	if (channel != tla2528->last_read_channel) {
> +		ret = tla2528_write_reg(client, TLA2528_CHANNEL_SEL_ADR, channel);
> +		if (ret < 0)
> +			return ret;
> +
> +		tla2528->last_read_channel = channel;
> +	}

If we implemented regmap with cache, then we could avoid having to
track last_read_channel. We could just call regmap_write() unconditionally
and the regmap framework would decide if it needs to actually do the write
or not.

> +
> +	ret = tla2528_read_sample(client);
> +	if (ret < 0)
> +		return ret;
> +
> +	*val = ret;
> +
> +	return 0;
> +}
> +
> +static int tla2528_read_raw(struct iio_dev *indio_dev,
> +			    struct iio_chan_spec const *chan,
> +			    int *val, int *val2, long mask)
> +{
> +	struct tla2528 *tla2528 = iio_priv(indio_dev);
> +	int ret;
> +
> +	switch (mask) {
> +	case IIO_CHAN_INFO_RAW:
> +		mutex_lock(&tla2528->lock);
> +		ret = tla2528_read(tla2528, chan->channel, val);
> +		mutex_unlock(&tla2528->lock);
> +		if (ret < 0)
> +			return ret;
> +
> +		return IIO_VAL_INT;
> +
> +	case IIO_CHAN_INFO_SCALE:
> +		*val = tla2528->vref_uv / 1000;

Why not just store vref_mV?

> +		*val2 = 12;
> +
> +		return IIO_VAL_FRACTIONAL_LOG2;
> +
> +	default:
> +		return -EINVAL;
> +	}
> +}
> +
> +#define TLA2528_CHAN(_chan, _name) { \
> +	.type = IIO_VOLTAGE,					\
> +	.channel = (_chan),					\
> +	.info_mask_separate = BIT(IIO_CHAN_INFO_RAW),		\
> +	.info_mask_shared_by_type = BIT(IIO_CHAN_INFO_SCALE),	\
> +	.datasheet_name = _name,				\
> +	.indexed = 1,						\
> +}
> +
> +static const struct iio_chan_spec tla2528_channel[] = {
> +	TLA2528_CHAN(0, "AIN0"),
> +	TLA2528_CHAN(1, "AIN1"),
> +	TLA2528_CHAN(2, "AIN2"),
> +	TLA2528_CHAN(3, "AIN3"),
> +	TLA2528_CHAN(4, "AIN4"),
> +	TLA2528_CHAN(5, "AIN5"),
> +	TLA2528_CHAN(6, "AIN6"),
> +	TLA2528_CHAN(7, "AIN7"),
> +};
> +
> +static const struct iio_info tla2528_info = {
> +	.read_raw = tla2528_read_raw,
> +};
> +
> +static int tla2528_probe(struct i2c_client *client)
> +{
> +	struct iio_dev *indio_dev;
> +	struct tla2528 *tla2528;
> +	int ret;
> +
> +	if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C |
> +				     I2C_FUNC_SMBUS_WRITE_WORD_DATA))
> +		return -EOPNOTSUPP;

Perhaps we should also fail if the adapter has I2C_AQ_NO_CLK_STRETCH?
It looks like clock stretching is required for the conversion time.

> +
> +	indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*tla2528));
> +	if (!indio_dev)
> +		return -ENOMEM;
> +
> +	tla2528 = iio_priv(indio_dev);

> +	i2c_set_clientdata(client, indio_dev);

Not sure this is needed if there is no i2c_get_clientdata() anywhere.

> +	tla2528->client = client;
> +

> +	indio_dev->name = client->name;

This should be the chip name ("tla2528"), not the I2C device name.

> +	indio_dev->info = &tla2528_info;
> +	indio_dev->modes = INDIO_DIRECT_MODE;
> +	indio_dev->channels = tla2528_channel;
> +	indio_dev->num_channels = ARRAY_SIZE(tla2528_channel);
> +
> +	mutex_init(&tla2528->lock);

Use dem_mutex_init().

> +
> +	tla2528->vref_uv = devm_regulator_get_enable_read_voltage(&client->dev,
> +								  "vref");
> +	if (tla2528->vref_uv < 0)
> +		return tla2528->vref_uv;
> +

> +	/* Set all inputs as analog */
> +	ret = tla2528_write_reg(tla2528->client, TLA2528_PIN_CFG_ADR, 0x00);
> +	if (ret < 0)
> +		return ret;
> +
> +	ret = tla2528_write_reg(tla2528->client, TLA2528_DATA_CFG_ADR,
> +				TLA2528_DATA_CFG_APPEND_STATUS);

Why? It doesn't appear to be used.

> +	if (ret < 0)
> +		return ret;
> +
> +	/* Set manual mode */
> +	ret = tla2528_write_reg(tla2528->client, TLA2528_SEQUENCE_CFG_ADR, 0x00);
> +	if (ret < 0)
> +		return ret;
> +


Perhaps simpler to just write the RST bit GENERAL_CFG to reset everything
to a known state?


> +	/* Init private data */
> +	tla2528->last_read_channel = ~0;
> +
> +	return devm_iio_device_register(&client->dev, indio_dev);
> +}
> +
> +static const struct i2c_device_id tla2528_id[] = {
> +	{ "tla2528", 0 },
> +	{ }
> +};
> +MODULE_DEVICE_TABLE(i2c, tla2528_id);
> +
> +static const struct of_device_id tla2528_of_match[] = {
> +	{ .compatible = "ti,tla2528", },
> +	{  },
> +};
> +MODULE_DEVICE_TABLE(of, tla2528_of_match);
> +
> +static struct i2c_driver tla2528_driver = {
> +	.driver = {
> +		.name = "tla2528",
> +		.of_match_table = tla2528_of_match,
> +	},
> +	.probe = tla2528_probe,
> +	.id_table = tla2528_id,
> +};
> +module_i2c_driver(tla2528_driver);
> +
> +MODULE_AUTHOR("Maxime Chevallier <maxime.chevallier@bootlin.com>");
> +MODULE_AUTHOR("Rodolfo Giometti <giometti@enneenne.com>");
> +MODULE_DESCRIPTION("Texas Instruments TLA2528 ADC driver");
> +MODULE_LICENSE("GPL");

Re: [PATCH 2/2] iio: adc: add driver for Texas Instruments TLA2528 adc

[Andy Shevchenko]

On Tue, Dec 23, 2025 at 11:33:02AM -0600, David Lechner wrote:
> On 12/23/25 11:12 AM, Maxime Chevallier wrote:
> > On 23/12/2025 16:55, Maxime Chevallier wrote:
> >> This adds a new driver for the TI TLA2528 ADC chip. It ha 8 12-bit
> >> channels, that can also be configured as 16-bit averaging channels.
> >>
> >> Add a very simple driver for it, allowing reading raw values for each
> >> channel.
> >>
> >> Signed-off-by: Rodolfo Giometti <giometti@enneenne.com>
> > 
> > Looking closer at this, Rodolfo hasn't seen this patch prior to me
> > sending it, so it should rather be :
> > 
> > Orginally-by: Rodolfo Giometti <giometti@enneenne.com>
> 
> I think the usual way would be to keep the Signed-off-by: and add
> Co-developed-by:.
> 
> See https://docs.kernel.org/process/submitting-patches.html#when-to-use-acked-by-cc-and-co-developed-by

Hmm... I think the Originally-by is used when the contributor gave
a PoC / basic idea in a form of code that was (heavily?) rewritten.

The only documentation mentions this tag is
Documentation/process/maintainer-tip.rst.

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH 2/2] iio: adc: add driver for Texas Instruments TLA2528 adc

[Jonathan Cameron]

On Tue, 23 Dec 2025 16:55:33 +0100
Maxime Chevallier <maxime.chevallier@bootlin.com> wrote:

> This adds a new driver for the TI TLA2528 ADC chip. It ha 8 12-bit
> channels, that can also be configured as 16-bit averaging channels.
> 
> Add a very simple driver for it, allowing reading raw values for each
> channel.
> 
> Signed-off-by: Rodolfo Giometti <giometti@enneenne.com>
> Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>

Hi Maxime,

A few extra bits from me

Thanks,

Jonathan


> diff --git a/drivers/iio/adc/ti-tla2528.c b/drivers/iio/adc/ti-tla2528.c
> new file mode 100644
> index 000000000000..9c572e730ffb
> --- /dev/null
> +++ b/drivers/iio/adc/ti-tla2528.c
> @@ -0,0 +1,209 @@

> +
> +static s32 tla2528_write_reg(const struct i2c_client *client, u8 reg, u8 val)
> +{
> +	u8 data[3] = {TLA2528_OP_WRITE_REG, reg, val};

Style wise. Prefer { TLA25... val };

- couple of spaces next to the brackets.

> +	int ret;
> +
> +	ret = i2c_master_send(client, data, 3);
> +
> +	return ret < 0 ? ret : 0;
> +}
> +
> +static int tla2528_read_sample(const struct i2c_client *client)
> +{
> +	__be16 data;
> +	int ret;
> +
> +	ret = i2c_master_recv(client, (char *)&data, 2);

sizeof(data)

> +	if (ret < 0)
> +		return ret;
> +
> +	return be16_to_cpu(data) >> 4;
> +}

> +
> +#define TLA2528_CHAN(_chan, _name) { \

The _ aren't adding anything here, so I'd drop them.

> +	.type = IIO_VOLTAGE,					\
> +	.channel = (_chan),					\
> +	.info_mask_separate = BIT(IIO_CHAN_INFO_RAW),		\
> +	.info_mask_shared_by_type = BIT(IIO_CHAN_INFO_SCALE),	\
> +	.datasheet_name = _name,				\
> +	.indexed = 1,						\
> +}

> +
> +static int tla2528_probe(struct i2c_client *client)
> +{
> +	struct iio_dev *indio_dev;
> +	struct tla2528 *tla2528;
> +	int ret;
> +
> +	if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C |
> +				     I2C_FUNC_SMBUS_WRITE_WORD_DATA))
> +		return -EOPNOTSUPP;
> +
> +	indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*tla2528));
> +	if (!indio_dev)
> +		return -ENOMEM;
> +
> +	tla2528 = iio_priv(indio_dev);
> +	i2c_set_clientdata(client, indio_dev);

Ever used? If not don't set it.

> +	tla2528->client = client;
> +
> +	indio_dev->name = client->name;

Prefer to see it hard coded as a string.  If we added extra firmware
types in future the content of client->name can become something other
than the part number.


> +	indio_dev->info = &tla2528_info;
> +	indio_dev->modes = INDIO_DIRECT_MODE;
> +	indio_dev->channels = tla2528_channel;
> +	indio_dev->num_channels = ARRAY_SIZE(tla2528_channel);
> +
> +	mutex_init(&tla2528->lock);
> +
> +	tla2528->vref_uv = devm_regulator_get_enable_read_voltage(&client->dev,
> +								  "vref");
> +	if (tla2528->vref_uv < 0)
> +		return tla2528->vref_uv;
> +
> +	/* Set all inputs as analog */
> +	ret = tla2528_write_reg(tla2528->client, TLA2528_PIN_CFG_ADR, 0x00);
> +	if (ret < 0)
> +		return ret;
> +
> +	ret = tla2528_write_reg(tla2528->client, TLA2528_DATA_CFG_ADR,
> +				TLA2528_DATA_CFG_APPEND_STATUS);
> +	if (ret < 0)
> +		return ret;
> +
> +	/* Set manual mode */
> +	ret = tla2528_write_reg(tla2528->client, TLA2528_SEQUENCE_CFG_ADR, 0x00);
> +	if (ret < 0)
> +		return ret;
> +
> +	/* Init private data */
> +	tla2528->last_read_channel = ~0;
> +
> +	return devm_iio_device_register(&client->dev, indio_dev);
> +}

Re: [PATCH 1/2] dt-bindings: iio: adc: add Texas Instruments TLA 2528

[Matti Vaittinen]

On 23/12/2025 20:26, David Lechner wrote:
> On 12/23/25 9:55 AM, Maxime Chevallier wrote:
>> The TI TLA 2528 is a simple 8 channel, 12-bit ADC? Add a binding
> 
> TLA2528 (no space). Also, why the "?"?
> 
>> documentation for it.
>>
>> Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
>> ---
>>   .../bindings/iio/adc/ti,tla2528.yaml          | 48 +++++++++++++++++++
>>   1 file changed, 48 insertions(+)
>>   create mode 100644 Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml
>>
>> diff --git a/Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml b/Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml
>> new file mode 100644
>> index 000000000000..0ee326d77014
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml
>> @@ -0,0 +1,48 @@
>> +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
>> +%YAML 1.2
>> +---
>> +$id: http://devicetree.org/schemas/iio/adc/ti,tla2528.yaml#
>> +$schema: http://devicetree.org/meta-schemas/core.yaml#
>> +
>> +title: Texas Instruments TLA2528 8-channel 12bit I2C ADC
> 
> 12-bit
> 
>> +
>> +maintainers:
>> +  - Maxime Chevallier <maxime.chevallier@bootlin.com>
>> +
>> +description: |
>> +  12bit 8-channel I2C ADC.
> 
> The title already says this. Either drop it or add new info.
> 
> Also, don't need the |.
> 
>> +
>> +properties:
>> +  compatible:
>> +    const: ti,tla2528
>> +
>> +  reg:
>> +    maxItems: 1
>> +
>> +  vref-supply:
>> +    description: Supply for 2.35V to 5.5V reference voltage
> 
> According the the datasheet, there are AVDD and DVDD supplies.
> Nothing named VREF or REF.
> 
> So instead:
> 
> avdd-supply: true
> dvdd-supply: true
> 
> 
> It looks like inputs can also be used as GPIOs, so
> 
> gpio-controller: true
> #gpio-cells:
>    const: 2
> 
> would be appropriate (it doesn't matter if the driver doesn't
> implement it, we know what the correct bindings are).
> 
>> +
>> +  "#io-channel-cells":
>> +    const: 1

I didn't check the data-sheet, but if the pins can be set to be GPIOs or 
ADC inputs, then I would require channels to be specified. It's only 8 
channels, so always listing channels that are present shouldn't be that 
big of a problem - and it should avoid one to add extra properties to 
denote channels used for GPIO if GPIOs need to be supported.

Well, I am not insisting this, there are folks that know this stuff 
better than I :)


Yours,
	-- Matti

---
Matti Vaittinen
Linux kernel developer at ROHM Semiconductors
Oulu Finland

~~ When things go utterly wrong vim users can always type :help! ~~

Re: [PATCH 2/2] iio: adc: add driver for Texas Instruments TLA2528 adc

[Matti Vaittinen]

On 23/12/2025 17:55, Maxime Chevallier wrote:
> This adds a new driver for the TI TLA2528 ADC chip. It ha 8 12-bit
> channels, that can also be configured as 16-bit averaging channels.
> 
> Add a very simple driver for it, allowing reading raw values for each
> channel.
> 
> Signed-off-by: Rodolfo Giometti <giometti@enneenne.com>
> Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
> ---
>   MAINTAINERS                  |   7 ++
>   drivers/iio/adc/Kconfig      |  10 ++
>   drivers/iio/adc/Makefile     |   1 +
>   drivers/iio/adc/ti-tla2528.c | 209 +++++++++++++++++++++++++++++++++++
>   4 files changed, 227 insertions(+)
>   create mode 100644 drivers/iio/adc/ti-tla2528.c
> 
> diff --git a/MAINTAINERS b/MAINTAINERS
> index dc731d37c8fe..5c382ae216c7 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -25866,6 +25866,13 @@ F:	include/dt-bindings/soc/ti,sci_pm_domain.h
>   F:	include/linux/soc/ti/ti_sci_inta_msi.h
>   F:	include/linux/soc/ti/ti_sci_protocol.h
>   
> +TEXAS INSTRUMENTS' TLA2528 ADC DRIVER
> +M:	Maxime Chevallier <maxime.chevallier@bootlin.com>
> +L:	linux-iio@vger.kernel.org
> +S:	Supported
> +F:	Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml
> +F:	drivers/iio/adc/ti-tla2528.c
> +
>   TEXAS INSTRUMENTS' TMP117 TEMPERATURE SENSOR DRIVER
>   M:	Puranjay Mohan <puranjay@kernel.org>
>   L:	linux-iio@vger.kernel.org
> diff --git a/drivers/iio/adc/Kconfig b/drivers/iio/adc/Kconfig
> index 58da8255525e..67376de410bf 100644

Hmm. Would it ease merging if MAINTAINERS changes were in their own patch?

> --- a/drivers/iio/adc/Kconfig
> +++ b/drivers/iio/adc/Kconfig
> @@ -1803,6 +1803,16 @@ config TI_LMP92064
>   	  This driver can also be built as a module. If so, the module will be called
>   	  ti-lmp92064.
>   
> +config TI_TLA2528
> +	tristate "Texas Instruments TLA2528 ADC driver"
> +	depends on I2C
> +	help
> +	  Say yes here to build support for Texas Instruments TLA2528
> +	  12-Bit 8-Channel ADC.
> +
> +	  To compile this driver as a module, choose M here: the module will be
> +	  called ti-tla2528.
> +
>   config TI_TLC4541
>   	tristate "Texas Instruments TLC4541 ADC driver"
>   	depends on SPI
> diff --git a/drivers/iio/adc/Makefile b/drivers/iio/adc/Makefile
> index 7cc8f9a12f76..941606defbf7 100644
> --- a/drivers/iio/adc/Makefile
> +++ b/drivers/iio/adc/Makefile
> @@ -157,6 +157,7 @@ obj-$(CONFIG_TI_ADS8344) += ti-ads8344.o
>   obj-$(CONFIG_TI_ADS8688) += ti-ads8688.o
>   obj-$(CONFIG_TI_AM335X_ADC) += ti_am335x_adc.o
>   obj-$(CONFIG_TI_LMP92064) += ti-lmp92064.o
> +obj-$(CONFIG_TI_TLA2528) += ti-tla2528.o
>   obj-$(CONFIG_TI_TLC4541) += ti-tlc4541.o
>   obj-$(CONFIG_TI_TSC2046) += ti-tsc2046.o
>   obj-$(CONFIG_TWL4030_MADC) += twl4030-madc.o
> diff --git a/drivers/iio/adc/ti-tla2528.c b/drivers/iio/adc/ti-tla2528.c
> new file mode 100644
> index 000000000000..9c572e730ffb
> --- /dev/null
> +++ b/drivers/iio/adc/ti-tla2528.c
> @@ -0,0 +1,209 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Driver for Texas Instruments TLA2528 ADC
> + *
> + * Copyright (C) 2020-2021 Rodolfo Giometti <giometti@enneenne.com>
> + * Copyright (C) 2025 Maxime Chevallier <maxime.chevallier@bootlin.com>
> + */
> +
> +#include <linux/delay.h>
> +#include <linux/i2c.h>
> +#include <linux/module.h>
> +#include <linux/mod_devicetable.h>
> +#include <linux/regulator/consumer.h>
> +
> +#include <linux/iio/iio.h>
> +
> +#define TLA2528_OP_WRITE_REG		0x08
> +
> +#define TLA2528_DATA_CFG_ADR		0x02
> +
> +/* Datasheet says [5:4] sets the append status, but only bit 4 is used */
> +#define TLA2528_DATA_CFG_APPEND_STATUS	BIT(4)
> +#define TLA2528_PIN_CFG_ADR		0x05
> +#define TLA2528_SEQUENCE_CFG_ADR	0x10
> +#define TLA2528_CHANNEL_SEL_ADR		0x11
> +
> +struct tla2528 {
> +	struct i2c_client *client;
> +	int vref_uv;
> +
> +	/* Protects manual channel selection, i.e. last_read_channel */
> +	struct mutex lock;
> +	u8 last_read_channel;
> +};
> +
> +static s32 tla2528_write_reg(const struct i2c_client *client, u8 reg, u8 val)
> +{
> +	u8 data[3] = {TLA2528_OP_WRITE_REG, reg, val};
> +	int ret;
> +
> +	ret = i2c_master_send(client, data, 3);
> +
> +	return ret < 0 ? ret : 0;
> +}
> +
> +static int tla2528_read_sample(const struct i2c_client *client)
> +{
> +	__be16 data;
> +	int ret;
> +
> +	ret = i2c_master_recv(client, (char *)&data, 2);
> +	if (ret < 0)
> +		return ret;
> +
> +	return be16_to_cpu(data) >> 4;
> +}

Can regmap be used for all the usual benefits?

> +
> +static int tla2528_read(struct tla2528 *tla2528, u8 channel, int *val)
> +{
> +	struct i2c_client *client = tla2528->client;
> +	int ret;
> +
> +	if (channel != tla2528->last_read_channel) {
> +		ret = tla2528_write_reg(client, TLA2528_CHANNEL_SEL_ADR, channel);
> +		if (ret < 0)
> +			return ret;
> +
> +		tla2528->last_read_channel = channel;
> +	}
> +
> +	ret = tla2528_read_sample(client);
> +	if (ret < 0)
> +		return ret;
> +
> +	*val = ret;
> +
> +	return 0;
> +}
> +
> +static int tla2528_read_raw(struct iio_dev *indio_dev,
> +			    struct iio_chan_spec const *chan,
> +			    int *val, int *val2, long mask)
> +{
> +	struct tla2528 *tla2528 = iio_priv(indio_dev);
> +	int ret;
> +
> +	switch (mask) {
> +	case IIO_CHAN_INFO_RAW:
> +		mutex_lock(&tla2528->lock);
> +		ret = tla2528_read(tla2528, chan->channel, val);
> +		mutex_unlock(&tla2528->lock);
> +		if (ret < 0)
> +			return ret;
> +
> +		return IIO_VAL_INT;
> +
> +	case IIO_CHAN_INFO_SCALE:
> +		*val = tla2528->vref_uv / 1000;
> +		*val2 = 12;
> +
> +		return IIO_VAL_FRACTIONAL_LOG2;
> +
> +	default:
> +		return -EINVAL;
> +	}
> +}
> +
> +#define TLA2528_CHAN(_chan, _name) { \
> +	.type = IIO_VOLTAGE,					\
> +	.channel = (_chan),					\
> +	.info_mask_separate = BIT(IIO_CHAN_INFO_RAW),		\
> +	.info_mask_shared_by_type = BIT(IIO_CHAN_INFO_SCALE),	\
> +	.datasheet_name = _name,				\
> +	.indexed = 1,						\
> +}
> +
> +static const struct iio_chan_spec tla2528_channel[] = {
> +	TLA2528_CHAN(0, "AIN0"),
> +	TLA2528_CHAN(1, "AIN1"),
> +	TLA2528_CHAN(2, "AIN2"),
> +	TLA2528_CHAN(3, "AIN3"),
> +	TLA2528_CHAN(4, "AIN4"),
> +	TLA2528_CHAN(5, "AIN5"),
> +	TLA2528_CHAN(6, "AIN6"),
> +	TLA2528_CHAN(7, "AIN7"),
> +};

Not really insisting a change here, merely giving a nudge :)

I wonder if the channels should be obtained from fwnode? I think the 
dt-review mentioned channel inputs may be used for GPIOs? One might be 
able to use the devm_iio_adc_device_alloc_chaninfo_se() to build the 
chan_spec based on the dt - depending on the non dt use-cases.

All in all, I think this is clean and nice driver.

> +static const struct iio_info tla2528_info = {
> +	.read_raw = tla2528_read_raw,
> +};
> +
> +static int tla2528_probe(struct i2c_client *client)
> +{
> +	struct iio_dev *indio_dev;
> +	struct tla2528 *tla2528;
> +	int ret;
> +
> +	if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C |
> +				     I2C_FUNC_SMBUS_WRITE_WORD_DATA))
> +		return -EOPNOTSUPP;
> +
> +	indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*tla2528));
> +	if (!indio_dev)
> +		return -ENOMEM;
> +
> +	tla2528 = iio_priv(indio_dev);
> +	i2c_set_clientdata(client, indio_dev);
> +	tla2528->client = client;
> +
> +	indio_dev->name = client->name;
> +	indio_dev->info = &tla2528_info;
> +	indio_dev->modes = INDIO_DIRECT_MODE;
> +	indio_dev->channels = tla2528_channel;
> +	indio_dev->num_channels = ARRAY_SIZE(tla2528_channel);
> +
> +	mutex_init(&tla2528->lock);
> +
> +	tla2528->vref_uv = devm_regulator_get_enable_read_voltage(&client->dev,
> +								  "vref");
> +	if (tla2528->vref_uv < 0)
> +		return tla2528->vref_uv;
> +
> +	/* Set all inputs as analog */
> +	ret = tla2528_write_reg(tla2528->client, TLA2528_PIN_CFG_ADR, 0x00);
> +	if (ret < 0)
> +		return ret;

As mentioned above, perhaps get the channels from the DT, and only mux 
the given channels?

> +
> +	ret = tla2528_write_reg(tla2528->client, TLA2528_DATA_CFG_ADR,
> +				TLA2528_DATA_CFG_APPEND_STATUS);
> +	if (ret < 0)
> +		return ret;
> +
> +	/* Set manual mode */
> +	ret = tla2528_write_reg(tla2528->client, TLA2528_SEQUENCE_CFG_ADR, 0x00);
> +	if (ret < 0)
> +		return ret;
> +
> +	/* Init private data */
> +	tla2528->last_read_channel = ~0;
> +
> +	return devm_iio_device_register(&client->dev, indio_dev);
> +}
> +
> +static const struct i2c_device_id tla2528_id[] = {
> +	{ "tla2528", 0 },
> +	{ }
> +};
> +MODULE_DEVICE_TABLE(i2c, tla2528_id);
> +
> +static const struct of_device_id tla2528_of_match[] = {
> +	{ .compatible = "ti,tla2528", },
> +	{  },
> +};
> +MODULE_DEVICE_TABLE(of, tla2528_of_match);
> +
> +static struct i2c_driver tla2528_driver = {
> +	.driver = {
> +		.name = "tla2528",
> +		.of_match_table = tla2528_of_match,
> +	},
> +	.probe = tla2528_probe,
> +	.id_table = tla2528_id,
> +};
> +module_i2c_driver(tla2528_driver);
> +
> +MODULE_AUTHOR("Maxime Chevallier <maxime.chevallier@bootlin.com>");
> +MODULE_AUTHOR("Rodolfo Giometti <giometti@enneenne.com>");
> +MODULE_DESCRIPTION("Texas Instruments TLA2528 ADC driver");
> +MODULE_LICENSE("GPL");

Yours,
	-- Matti

-- 
---
Matti Vaittinen
Linux kernel developer at ROHM Semiconductors
Oulu Finland

~~ When things go utterly wrong vim users can always type :help! ~~

Re: [PATCH 1/2] dt-bindings: iio: adc: add Texas Instruments TLA 2528

[Andy Shevchenko]

On Mon, Dec 29, 2025 at 10:04 AM Matti Vaittinen
<mazziesaccount@gmail.com> wrote:
> On 23/12/2025 20:26, David Lechner wrote:
> > On 12/23/25 9:55 AM, Maxime Chevallier wrote:

...

> > It looks like inputs can also be used as GPIOs, so
> >
> > gpio-controller: true
> > #gpio-cells:
> >    const: 2
> >
> > would be appropriate (it doesn't matter if the driver doesn't
> > implement it, we know what the correct bindings are).
> >
> >> +
> >> +  "#io-channel-cells":
> >> +    const: 1
>
> I didn't check the data-sheet, but if the pins can be set to be GPIOs or
> ADC inputs, then I would require channels to be specified. It's only 8
> channels, so always listing channels that are present shouldn't be that
> big of a problem - and it should avoid one to add extra properties to
> denote channels used for GPIO if GPIOs need to be supported.
>
> Well, I am not insisting this, there are folks that know this stuff
> better than I :)

Why would we need an extra property for that? GPIO controller has a
property for valid_mask, should be enough to handle this case, no?

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH 2/2] iio: adc: add driver for Texas Instruments TLA2528 adc

[Andy Shevchenko]

On Tue, Dec 23, 2025 at 5:55 PM Maxime Chevallier
<maxime.chevallier@bootlin.com> wrote:
>
> This adds a new driver for the TI TLA2528 ADC chip. It ha 8 12-bit
> channels, that can also be configured as 16-bit averaging channels.
>
> Add a very simple driver for it, allowing reading raw values for each
> channel.

...

> +#include <linux/delay.h>
> +#include <linux/i2c.h>
> +#include <linux/module.h>
> +#include <linux/mod_devicetable.h>
> +#include <linux/regulator/consumer.h>

Follow IWYU. Here are missing headers, such as bits.h, mutex.h, types.h.

...

> +       case IIO_CHAN_INFO_SCALE:
> +               *val = tla2528->vref_uv / 1000;

(MICRO/MILLI) ?

> +               *val2 = 12;
> +
> +               return IIO_VAL_FRACTIONAL_LOG2;



> +       tla2528->vref_uv = devm_regulator_get_enable_read_voltage(&client->dev,
> +                                                                 "vref");

With

  struct device *dev = &client->dev;

at the top of the function this will be one line and others also can
be shortened.

> +       if (tla2528->vref_uv < 0)
> +               return tla2528->vref_uv;

...

> +       /* Set all inputs as analog */
> +       ret = tla2528_write_reg(tla2528->client, TLA2528_PIN_CFG_ADR, 0x00);

Why not simply use the "client"?

> +       if (ret < 0)
> +               return ret;
> +
> +       ret = tla2528_write_reg(tla2528->client, TLA2528_DATA_CFG_ADR,
> +                               TLA2528_DATA_CFG_APPEND_STATUS);
> +       if (ret < 0)
> +               return ret;
> +
> +       /* Set manual mode */
> +       ret = tla2528_write_reg(tla2528->client, TLA2528_SEQUENCE_CFG_ADR, 0x00);
> +       if (ret < 0)
> +               return ret;

Ditto.

...

> +static const struct i2c_device_id tla2528_id[] = {
> +       { "tla2528", 0 },

No ', 0' part.

> +       { }
> +};

...

> +static const struct of_device_id tla2528_of_match[] = {
> +       { .compatible = "ti,tla2528", },

No inner comma.

> +       {  },

No trailing comma.

> +};

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH 1/2] dt-bindings: iio: adc: add Texas Instruments TLA 2528

[Matti Vaittinen]

On 29/12/2025 11:31, Andy Shevchenko wrote:
> On Mon, Dec 29, 2025 at 10:04 AM Matti Vaittinen
> <mazziesaccount@gmail.com> wrote:
>> On 23/12/2025 20:26, David Lechner wrote:
>>> On 12/23/25 9:55 AM, Maxime Chevallier wrote:
> 
> ...
> 
>>> It looks like inputs can also be used as GPIOs, so
>>>
>>> gpio-controller: true
>>> #gpio-cells:
>>>     const: 2
>>>
>>> would be appropriate (it doesn't matter if the driver doesn't
>>> implement it, we know what the correct bindings are).
>>>
>>>> +
>>>> +  "#io-channel-cells":
>>>> +    const: 1
>>
>> I didn't check the data-sheet, but if the pins can be set to be GPIOs or
>> ADC inputs, then I would require channels to be specified. It's only 8
>> channels, so always listing channels that are present shouldn't be that
>> big of a problem - and it should avoid one to add extra properties to
>> denote channels used for GPIO if GPIOs need to be supported.
>>
>> Well, I am not insisting this, there are folks that know this stuff
>> better than I :)
> 
> Why would we need an extra property for that? GPIO controller has a
> property for valid_mask, should be enough to handle this case, no?
> 
Ah. You're right. The "valid_mask" should be perfectly usable.

I might still require the channel information to make it explicit - but 
as I said, I leave this for others to decide :)

Yours,
	-- Matti


---
Matti Vaittinen
Linux kernel developer at ROHM Semiconductors
Oulu Finland

~~ When things go utterly wrong vim users can always type :help! ~~

Re: [PATCH 2/2] iio: adc: add driver for Texas Instruments TLA2528 adc

[Jonathan Cameron]

On Mon, 29 Dec 2025 10:20:23 +0200
Matti Vaittinen <mazziesaccount@gmail.com> wrote:

> On 23/12/2025 17:55, Maxime Chevallier wrote:
> > This adds a new driver for the TI TLA2528 ADC chip. It ha 8 12-bit
> > channels, that can also be configured as 16-bit averaging channels.
> > 
> > Add a very simple driver for it, allowing reading raw values for each
> > channel.
> > 
> > Signed-off-by: Rodolfo Giometti <giometti@enneenne.com>
> > Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
> > ---
> >   MAINTAINERS                  |   7 ++
> >   drivers/iio/adc/Kconfig      |  10 ++
> >   drivers/iio/adc/Makefile     |   1 +
> >   drivers/iio/adc/ti-tla2528.c | 209 +++++++++++++++++++++++++++++++++++
> >   4 files changed, 227 insertions(+)
> >   create mode 100644 drivers/iio/adc/ti-tla2528.c
> > 
> > diff --git a/MAINTAINERS b/MAINTAINERS
> > index dc731d37c8fe..5c382ae216c7 100644
> > --- a/MAINTAINERS
> > +++ b/MAINTAINERS
> > @@ -25866,6 +25866,13 @@ F:	include/dt-bindings/soc/ti,sci_pm_domain.h
> >   F:	include/linux/soc/ti/ti_sci_inta_msi.h
> >   F:	include/linux/soc/ti/ti_sci_protocol.h
> >   
> > +TEXAS INSTRUMENTS' TLA2528 ADC DRIVER
> > +M:	Maxime Chevallier <maxime.chevallier@bootlin.com>
> > +L:	linux-iio@vger.kernel.org
> > +S:	Supported
> > +F:	Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml
> > +F:	drivers/iio/adc/ti-tla2528.c
> > +
> >   TEXAS INSTRUMENTS' TMP117 TEMPERATURE SENSOR DRIVER
> >   M:	Puranjay Mohan <puranjay@kernel.org>
> >   L:	linux-iio@vger.kernel.org
> > diff --git a/drivers/iio/adc/Kconfig b/drivers/iio/adc/Kconfig
> > index 58da8255525e..67376de410bf 100644  
> 
> Hmm. Would it ease merging if MAINTAINERS changes were in their own patch?

Not particularly.  Though I personally slightly prefer the logic
of bringing the entry in with the first file, then adding additional files
in later patches.

Given it is huge and in alphabetical order, conflicts in MAINTAINERS are
fairly rare and trivial to resolve.

Re: [PATCH 2/2] iio: adc: add driver for Texas Instruments TLA2528 adc

[Matti Vaittinen]

On 31/12/2025 19:12, Jonathan Cameron wrote:
> On Mon, 29 Dec 2025 10:20:23 +0200
> Matti Vaittinen <mazziesaccount@gmail.com> wrote:
> 
>> On 23/12/2025 17:55, Maxime Chevallier wrote:
>>> This adds a new driver for the TI TLA2528 ADC chip. It ha 8 12-bit
>>> channels, that can also be configured as 16-bit averaging channels.
>>>
>>> Add a very simple driver for it, allowing reading raw values for each
>>> channel.
>>>
>>> Signed-off-by: Rodolfo Giometti <giometti@enneenne.com>
>>> Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
>>> ---
>>>    MAINTAINERS                  |   7 ++
>>>    drivers/iio/adc/Kconfig      |  10 ++
>>>    drivers/iio/adc/Makefile     |   1 +
>>>    drivers/iio/adc/ti-tla2528.c | 209 +++++++++++++++++++++++++++++++++++
>>>    4 files changed, 227 insertions(+)
>>>    create mode 100644 drivers/iio/adc/ti-tla2528.c
>>>
>>> diff --git a/MAINTAINERS b/MAINTAINERS
>>> index dc731d37c8fe..5c382ae216c7 100644
>>> --- a/MAINTAINERS
>>> +++ b/MAINTAINERS
>>> @@ -25866,6 +25866,13 @@ F:	include/dt-bindings/soc/ti,sci_pm_domain.h
>>>    F:	include/linux/soc/ti/ti_sci_inta_msi.h
>>>    F:	include/linux/soc/ti/ti_sci_protocol.h
>>>    
>>> +TEXAS INSTRUMENTS' TLA2528 ADC DRIVER
>>> +M:	Maxime Chevallier <maxime.chevallier@bootlin.com>
>>> +L:	linux-iio@vger.kernel.org
>>> +S:	Supported
>>> +F:	Documentation/devicetree/bindings/iio/adc/ti,tla2528.yaml
>>> +F:	drivers/iio/adc/ti-tla2528.c
>>> +
>>>    TEXAS INSTRUMENTS' TMP117 TEMPERATURE SENSOR DRIVER
>>>    M:	Puranjay Mohan <puranjay@kernel.org>
>>>    L:	linux-iio@vger.kernel.org
>>> diff --git a/drivers/iio/adc/Kconfig b/drivers/iio/adc/Kconfig
>>> index 58da8255525e..67376de410bf 100644
>>
>> Hmm. Would it ease merging if MAINTAINERS changes were in their own patch?
> 
> Not particularly.  Though I personally slightly prefer the logic
> of bringing the entry in with the first file, then adding additional files
> in later patches.
> 
> Given it is huge and in alphabetical order, conflicts in MAINTAINERS are
> fairly rare and trivial to resolve.
> 

Thanks for this clarification :)

I don't know where I had picked up this idea, but I thought that the 
volume of changes in MAINTAINERs was somewhat annoying source of 
conflicts. I sit and type corrected :)


Yours,
	-- Matti

---
Matti Vaittinen
Linux kernel developer at ROHM Semiconductors
Oulu Finland

~~ When things go utterly wrong vim users can always type :help! ~~