[PATCH 0/3] staging: iio: accel: adis16203: fix axis mapping and cleanups

[PATCH 0/3] staging: iio: accel: adis16203: fix axis mapping and cleanups

[Abdelnasser Hussein]

Hi all,

This patch series addresses a fundamental hardware mapping error in the
ADIS16203 driver and performs a few necessary cleanups.

The ADIS16203 is a single-axis inclinometer. The current driver mistakenly
exposes two separate axes (X and Y), leaving a FIXME comment on the Y-axis.
According to the datasheet, these are actually just two different output
formats for the same axis (0 to 360 degrees, and +/- 180 degrees).

This series removes the unsupported Y-axis, fixes the inclination naming,
adds a missing SPI device ID table to ensure proper module autoloading,
and cleans up the register macros to align with kernel coding style.

Best regards,
Abdelnasser Hussein

Abdelnasser Hussein (3):
  staging: iio: accel: adis16203: fix inclination naming and remove
    unsupported Y-axis
  staging: iio: accel: adis16203: add missing spi_device_id table
  staging: iio: accel: adis16203: clean up register definitions and drop
    redundant comments

 drivers/staging/iio/accel/adis16203.c | 108 ++++++--------------------
 1 file changed, 24 insertions(+), 84 deletions(-)

-- 
2.54.0

[PATCH 1/3] staging: iio: accel: adis16203: fix inclination naming and remove unsupported Y-axis

[Abdelnasser Hussein]

The ADIS16203 measures inclination around a single axis. As per the
datasheet, register 0x0C gives a 360-degree output, and 0x0E gives a
+/- 180-degree output of the exact same measurement.

The current driver incorrectly maps 0x0E as a separate Y-axis. This patch
removes the phantom Y-axis, updates the naming, and adjusts the timestamp
channel index.

Note: The ADIS_INCLI_CHAN(X, ...) modifier is intentionally kept to
preserve the existing userspace ABI (incli_x_raw). Furthermore, removing
the phantom Y-axis inherently shifts the scan indexes of subsequent
channels down by one, which is expected to correct the hardware
representation.

Fixes: 2df79e1bddc4 ("staging: iio: accel: Rename source files and update Makefile")
Signed-off-by: Abdelnasser Hussein <abdelnasserhussein11@gmail.com>
---
 drivers/staging/iio/accel/adis16203.c | 13 +++----------
 1 file changed, 3 insertions(+), 10 deletions(-)

diff --git a/drivers/staging/iio/accel/adis16203.c b/drivers/staging/iio/accel/adis16203.c
index 830ff38fda92..eddb786d4747 100644
--- a/drivers/staging/iio/accel/adis16203.c
+++ b/drivers/staging/iio/accel/adis16203.c
@@ -29,10 +29,7 @@
 #define ADIS16203_TEMP_OUT       0x0A
 
 /* Output, x-axis inclination */
-#define ADIS16203_XINCL_OUT      0x0C
-
-/* Output, y-axis inclination */
-#define ADIS16203_YINCL_OUT      0x0E
+#define ADIS16203_INCL_OUT       0x0C
 
 /* Incline null calibration */
 #define ADIS16203_INCL_NULL      0x18
@@ -129,7 +126,6 @@
 
 enum adis16203_scan {
 	 ADIS16203_SCAN_INCLI_X,
-	 ADIS16203_SCAN_INCLI_Y,
 	 ADIS16203_SCAN_SUPPLY,
 	 ADIS16203_SCAN_AUX_ADC,
 	 ADIS16203_SCAN_TEMP,
@@ -208,13 +204,10 @@ static int adis16203_read_raw(struct iio_dev *indio_dev,
 static const struct iio_chan_spec adis16203_channels[] = {
 	ADIS_SUPPLY_CHAN(ADIS16203_SUPPLY_OUT, ADIS16203_SCAN_SUPPLY, 0, 12),
 	ADIS_AUX_ADC_CHAN(ADIS16203_AUX_ADC, ADIS16203_SCAN_AUX_ADC, 0, 12),
-	ADIS_INCLI_CHAN(X, ADIS16203_XINCL_OUT, ADIS16203_SCAN_INCLI_X,
+	ADIS_INCLI_CHAN(X, ADIS16203_INCL_OUT, ADIS16203_SCAN_INCLI_X,
 			BIT(IIO_CHAN_INFO_CALIBBIAS), 0, 14),
-	/* Fixme: Not what it appears to be - see data sheet */
-	ADIS_INCLI_CHAN(Y, ADIS16203_YINCL_OUT, ADIS16203_SCAN_INCLI_Y,
-			0, 0, 14),
 	ADIS_TEMP_CHAN(ADIS16203_TEMP_OUT, ADIS16203_SCAN_TEMP, 0, 12),
-	IIO_CHAN_SOFT_TIMESTAMP(5),
+	IIO_CHAN_SOFT_TIMESTAMP(4),
 };
 
 static const struct iio_info adis16203_info = {
-- 
2.54.0

Re: [PATCH 1/3] staging: iio: accel: adis16203: fix inclination naming and remove unsupported Y-axis

[Joshua Crofts]

On Tue, 29 Sep 2026 21:48:58 +0300
Abdelnasser Hussein <abdelnasserhussein11@gmail.com> wrote:

> The ADIS16203 measures inclination around a single axis. As per the
> datasheet, register 0x0C gives a 360-degree output, and 0x0E gives a
> +/- 180-degree output of the exact same measurement.
> 
> The current driver incorrectly maps 0x0E as a separate Y-axis. This patch
> removes the phantom Y-axis, updates the naming, and adjusts the timestamp
> channel index.
> 
> Note: The ADIS_INCLI_CHAN(X, ...) modifier is intentionally kept to
> preserve the existing userspace ABI (incli_x_raw). Furthermore, removing
> the phantom Y-axis inherently shifts the scan indexes of subsequent
> channels down by one, which is expected to correct the hardware
> representation.
> 
> Fixes: 2df79e1bddc4 ("staging: iio: accel: Rename source files and update Makefile")
> Signed-off-by: Abdelnasser Hussein <abdelnasserhussein11@gmail.com>
> ---

Reviewed-by: Joshua Crofts <joshua.crofts1@gmail.com>

-- 
Kind regards,
Joshua Crofts

Re: [PATCH 1/3] staging: iio: accel: adis16203: fix inclination naming and remove unsupported Y-axis

[Andy Shevchenko]

On Tue, Sep 29, 2026 at 09:48:58PM +0300, Abdelnasser Hussein wrote:
> The ADIS16203 measures inclination around a single axis. As per the
> datasheet, register 0x0C gives a 360-degree output, and 0x0E gives a
> +/- 180-degree output of the exact same measurement.

> The current driver incorrectly maps 0x0E as a separate Y-axis. This patch
> removes the phantom Y-axis, updates the naming, and adjusts the timestamp
> channel index.

Read Submitting Patches about imperative mode (search for "This patch").

> Note: The ADIS_INCLI_CHAN(X, ...) modifier is intentionally kept to
> preserve the existing userspace ABI (incli_x_raw).

This is a staging driver, why do we care about ABI preservations?
Yes, ideally it would be good to have existing user space software
continue working as before, but if this adds a confusion, better to
rework an ABI as well.

> Furthermore, removing
> the phantom Y-axis inherently shifts the scan indexes of subsequent
> channels down by one, which is expected to correct the hardware
> representation.

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH 1/3] staging: iio: accel: adis16203: fix inclination naming and remove unsupported Y-axis

[nasser]

Hi Andy

Thank you for the review.

> Read Submitting Patches about imperative mode (search for "This patch").

I will reword the commit message to  use the imperative mood in v2.

> This is a staging driver, why do we care about ABI preservations?
> Yes, ideally it would be good to have existing user space software
> continue working as before, but if this adds a confusion, better to
> rework an ABI as well.

 I will completely drop the 'X' modifier from the channel definition in v2.

Best regards,
Abdelnasser

[PATCH 2/3] staging: iio: accel: adis16203: add missing spi_device_id table

[Abdelnasser Hussein]

The driver currently relies on OF/DT matching but lacks an SPI device
ID table. This prevents module auto-loading on systems without device
tree support.

Add the spi_device_id table to resolve this and ensure the driver loads
properly across all supported platforms.

Signed-off-by: Abdelnasser Hussein <abdelnasserhussein11@gmail.com>
---
 drivers/staging/iio/accel/adis16203.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/staging/iio/accel/adis16203.c b/drivers/staging/iio/accel/adis16203.c
index eddb786d4747..905485fd556f 100644
--- a/drivers/staging/iio/accel/adis16203.c
+++ b/drivers/staging/iio/accel/adis16203.c
@@ -11,6 +11,7 @@
 #include <linux/iio/imu/adis.h>
 
 #include <linux/kernel.h>
+#include <linux/mod_devicetable.h>
 #include <linux/module.h>
 #include <linux/spi/spi.h>
 
@@ -292,12 +293,19 @@ static const struct of_device_id adis16203_of_match[] = {
 
 MODULE_DEVICE_TABLE(of, adis16203_of_match);
 
+static const struct spi_device_id adis16203_id[] = {
+	{ "adis16203", 0 },
+	{ }
+};
+MODULE_DEVICE_TABLE(spi, adis16203_id);
+
 static struct spi_driver adis16203_driver = {
 	.driver = {
 		.name = "adis16203",
 		.of_match_table = adis16203_of_match,
 	},
 	.probe = adis16203_probe,
+	.id_table = adis16203_id,
 };
 module_spi_driver(adis16203_driver);
 
-- 
2.54.0

Re: [PATCH 2/3] staging: iio: accel: adis16203: add missing spi_device_id table

[Joshua Crofts]

On Tue, 29 Sep 2026 21:48:59 +0300
Abdelnasser Hussein <abdelnasserhussein11@gmail.com> wrote:

> The driver currently relies on OF/DT matching but lacks an SPI device
> ID table. This prevents module auto-loading on systems without device
> tree support.
> 
> Add the spi_device_id table to resolve this and ensure the driver loads
> properly across all supported platforms.
> 
> Signed-off-by: Abdelnasser Hussein <abdelnasserhussein11@gmail.com>
> ---
>  drivers/staging/iio/accel/adis16203.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/drivers/staging/iio/accel/adis16203.c b/drivers/staging/iio/accel/adis16203.c
> index eddb786d4747..905485fd556f 100644
> --- a/drivers/staging/iio/accel/adis16203.c
> +++ b/drivers/staging/iio/accel/adis16203.c
> @@ -11,6 +11,7 @@
>  #include <linux/iio/imu/adis.h>
>  
>  #include <linux/kernel.h>
> +#include <linux/mod_devicetable.h>

No need for mod_devicetable.h, this is imported using spi.h

>  #include <linux/module.h>
>  #include <linux/spi/spi.h>
>  
> @@ -292,12 +293,19 @@ static const struct of_device_id adis16203_of_match[] = {
>  
>  MODULE_DEVICE_TABLE(of, adis16203_of_match);
>  
> +static const struct spi_device_id adis16203_id[] = {
> +	{ "adis16203", 0 },

C99 initializers please.

> +	{ }
> +};
> +MODULE_DEVICE_TABLE(spi, adis16203_id);
> +
>  static struct spi_driver adis16203_driver = {
>  	.driver = {
>  		.name = "adis16203",
>  		.of_match_table = adis16203_of_match,
>  	},
>  	.probe = adis16203_probe,
> +	.id_table = adis16203_id,
>  };
>  module_spi_driver(adis16203_driver);
>  

Additionally, you should delete the MODULE_ALIAS("spi:adis16203") line.

-- 
Kind regards,
Joshua Crofts

Re: [PATCH 2/3] staging: iio: accel: adis16203: add missing spi_device_id table

[nasser]

Hi Joshua,

Thank you for taking the time to review this.

> No need for mod_devicetable.h, this is imported using spi.h
> C99 initializers please.
> Additionally, you should delete the MODULE_ALIAS("spi:adis16203") line.

I will drop the unnecessary header, update the spi_device_id array to
use C99 initializers, and remove the redundant MODULE_ALIAS in the
upcoming v2.

Best regards,
Abdelnasser

[PATCH 3/3] staging: iio: accel: adis16203: clean up register definitions and drop redundant comments

[Abdelnasser Hussein]

Drop redundant, single-line comments above register macros that provide
no additional value.

Vertically align the macro values with tabs to improve readability and
conform to standard kernel coding style.

Signed-off-by: Abdelnasser Hussein <abdelnasserhussein11@gmail.com>
---
 drivers/staging/iio/accel/adis16203.c | 87 ++++-----------------------
 1 file changed, 13 insertions(+), 74 deletions(-)

diff --git a/drivers/staging/iio/accel/adis16203.c b/drivers/staging/iio/accel/adis16203.c
index 905485fd556f..b0fa2701cbd1 100644
--- a/drivers/staging/iio/accel/adis16203.c
+++ b/drivers/staging/iio/accel/adis16203.c
@@ -17,105 +17,44 @@
 
 #define ADIS16203_STARTUP_DELAY 220 /* ms */
 
-/* Flash memory write count */
 #define ADIS16203_FLASH_CNT      0x00
-
-/* Output, power supply */
 #define ADIS16203_SUPPLY_OUT     0x02
-
-/* Output, auxiliary ADC input */
 #define ADIS16203_AUX_ADC        0x08
-
-/* Output, temperature */
 #define ADIS16203_TEMP_OUT       0x0A
-
-/* Output, x-axis inclination */
 #define ADIS16203_INCL_OUT       0x0C
-
-/* Incline null calibration */
 #define ADIS16203_INCL_NULL      0x18
-
-/* Alarm 1 amplitude threshold */
 #define ADIS16203_ALM_MAG1       0x20
-
-/* Alarm 2 amplitude threshold */
 #define ADIS16203_ALM_MAG2       0x22
-
-/* Alarm 1, sample period */
 #define ADIS16203_ALM_SMPL1      0x24
-
-/* Alarm 2, sample period */
 #define ADIS16203_ALM_SMPL2      0x26
-
-/* Alarm control */
 #define ADIS16203_ALM_CTRL       0x28
-
-/* Auxiliary DAC data */
 #define ADIS16203_AUX_DAC        0x30
-
-/* General-purpose digital input/output control */
 #define ADIS16203_GPIO_CTRL      0x32
-
-/* Miscellaneous control */
 #define ADIS16203_MSC_CTRL       0x34
-
-/* Internal sample period (rate) control */
 #define ADIS16203_SMPL_PRD       0x36
-
-/* Operation, filter configuration */
 #define ADIS16203_AVG_CNT        0x38
-
-/* Operation, sleep mode control */
 #define ADIS16203_SLP_CNT        0x3A
-
-/* Diagnostics, system status register */
 #define ADIS16203_DIAG_STAT      0x3C
-
-/* Operation, system command register */
 #define ADIS16203_GLOB_CMD       0x3E
 
 /* MSC_CTRL */
 
-/* Self-test at power-on: 1 = disabled, 0 = enabled */
-#define ADIS16203_MSC_CTRL_PWRUP_SELF_TEST      BIT(10)
-
-/* Reverses rotation of both inclination outputs */
-#define ADIS16203_MSC_CTRL_REVERSE_ROT_EN       BIT(9)
-
-/* Self-test enable */
-#define ADIS16203_MSC_CTRL_SELF_TEST_EN         BIT(8)
-
-/* Data-ready enable: 1 = enabled, 0 = disabled */
-#define ADIS16203_MSC_CTRL_DATA_RDY_EN          BIT(2)
-
-/* Data-ready polarity: 1 = active high, 0 = active low */
-#define ADIS16203_MSC_CTRL_ACTIVE_HIGH          BIT(1)
-
-/* Data-ready line selection: 1 = DIO1, 0 = DIO0 */
-#define ADIS16203_MSC_CTRL_DATA_RDY_DIO1        BIT(0)
+#define ADIS16203_MSC_CTRL_PWRUP_SELF_TEST	BIT(10)
+#define ADIS16203_MSC_CTRL_REVERSE_ROT_EN	BIT(9)
+#define ADIS16203_MSC_CTRL_SELF_TEST_EN	BIT(8)
+#define ADIS16203_MSC_CTRL_DATA_RDY_EN		BIT(2)
+#define ADIS16203_MSC_CTRL_ACTIVE_HIGH		BIT(1)
+#define ADIS16203_MSC_CTRL_DATA_RDY_DIO1	BIT(0)
 
 /* DIAG_STAT */
 
-/* Alarm 2 status: 1 = alarm active, 0 = alarm inactive */
-#define ADIS16203_DIAG_STAT_ALARM2        BIT(9)
-
-/* Alarm 1 status: 1 = alarm active, 0 = alarm inactive */
-#define ADIS16203_DIAG_STAT_ALARM1        BIT(8)
-
-/* Self-test diagnostic error flag */
-#define ADIS16203_DIAG_STAT_SELFTEST_FAIL_BIT 5
-
-/* SPI communications failure */
-#define ADIS16203_DIAG_STAT_SPI_FAIL_BIT      3
-
-/* Flash update failure */
-#define ADIS16203_DIAG_STAT_FLASH_UPT_BIT     2
-
-/* Power supply above 3.625 V */
-#define ADIS16203_DIAG_STAT_POWER_HIGH_BIT    1
-
-/* Power supply below 2.975 V */
-#define ADIS16203_DIAG_STAT_POWER_LOW_BIT     0
+#define ADIS16203_DIAG_STAT_ALARM2		BIT(9)
+#define ADIS16203_DIAG_STAT_ALARM1		BIT(8)
+#define ADIS16203_DIAG_STAT_SELFTEST_FAIL_BIT	5
+#define ADIS16203_DIAG_STAT_SPI_FAIL_BIT	3
+#define ADIS16203_DIAG_STAT_FLASH_UPT_BIT	2
+#define ADIS16203_DIAG_STAT_POWER_HIGH_BIT	1
+#define ADIS16203_DIAG_STAT_POWER_LOW_BIT	0
 
 /* GLOB_CMD */
 
-- 
2.54.0

Re: [PATCH 3/3] staging: iio: accel: adis16203: clean up register definitions and drop redundant comments

[Joshua Crofts]

On Tue, 29 Sep 2026 21:49:00 +0300
Abdelnasser Hussein <abdelnasserhussein11@gmail.com> wrote:

> Drop redundant, single-line comments above register macros that provide
> no additional value.
> 
> Vertically align the macro values with tabs to improve readability and
> conform to standard kernel coding style.
> 
> Signed-off-by: Abdelnasser Hussein <abdelnasserhussein11@gmail.com>
> ---

This seems okay, at least the macros are named exactly like the registers
in the datasheet... but I wonder what the others think about this.

-- 
Kind regards,
Joshua Crofts