[PATCH 0/4] iio: magnetometer: ak8975: modernize and cleanup driver

[PATCH 0/4] iio: magnetometer: ak8975: modernize and cleanup driver

[Joshua Crofts]

This series modernizes and cleans up the ak8975 driver, migrating to
more modern kernel APIs and cleaning up legacy code.

Changes include:
 - replacing usleep_range() with fsleep() for optimization
 - migrating to iopoll macros for better device polling
 - removing unused headers and adding new ones
 - checkpatch.pl fixes

Joshua Crofts (4):
  iio: magnetometer: ak8975: header cleanup
  iio: magnetometer: ak8975: replace usleep_range() with fsleep()
  iio: magnetometer: ak8975: change 'u8*' to 'u8 *' in cast
  iio: magnetometer: ak8975: modernize polling loops with iopoll()
    macros

 drivers/iio/magnetometer/ak8975.c | 77 ++++++++++++++++---------------
 1 file changed, 39 insertions(+), 38 deletions(-)

-- 
2.47.3

[PATCH 1/4] iio: magnetometer: ak8975: header cleanup

[Joshua Crofts]

Clean up headers by removing proxy kernel.h header and adding new
headers to ensure atomicity (array_size.h, dev_printk.h, types.h,
asm/byteorder.h). Removed unused headers (slab.h, iio/sysfs.h,
iio/trigger.h) and added minmax.h, property.h, wait.h, irqreturn.h to
mitigate transient dependencies during compilation.

Audited using the include-what-you-use tool.

Signed-off-by: Joshua Crofts <joshua.crofts1@gmail.com>
---
 drivers/iio/magnetometer/ak8975.c | 27 ++++++++++++++++-----------
 1 file changed, 16 insertions(+), 11 deletions(-)

diff --git a/drivers/iio/magnetometer/ak8975.c b/drivers/iio/magnetometer/ak8975.c
index 44782c26698..f45dca0a6f0 100644
--- a/drivers/iio/magnetometer/ak8975.c
+++ b/drivers/iio/magnetometer/ak8975.c
@@ -7,24 +7,29 @@
  * Copyright (c) 2010, NVIDIA Corporation.
  */
 
-#include <linux/module.h>
-#include <linux/mod_devicetable.h>
-#include <linux/kernel.h>
-#include <linux/slab.h>
+#include <linux/array_size.h>
+#include <linux/bitops.h>
+#include <linux/delay.h>
+#include <linux/dev_printk.h>
+#include <linux/err.h>
+#include <linux/gpio/consumer.h>
 #include <linux/i2c.h>
 #include <linux/interrupt.h>
-#include <linux/err.h>
+#include <linux/irqreturn.h>
+#include <linux/minmax.h>
+#include <linux/mod_devicetable.h>
+#include <linux/module.h>
 #include <linux/mutex.h>
-#include <linux/delay.h>
-#include <linux/bitops.h>
-#include <linux/gpio/consumer.h>
-#include <linux/regulator/consumer.h>
 #include <linux/pm_runtime.h>
+#include <linux/property.h>
+#include <linux/regulator/consumer.h>
+#include <linux/types.h>
+#include <linux/wait.h>
+
+#include <asm/byteorder.h>
 
 #include <linux/iio/iio.h>
-#include <linux/iio/sysfs.h>
 #include <linux/iio/buffer.h>
-#include <linux/iio/trigger.h>
 #include <linux/iio/trigger_consumer.h>
 #include <linux/iio/triggered_buffer.h>
 
-- 
2.47.3

Re: [PATCH 1/4] iio: magnetometer: ak8975: header cleanup

[Andy Shevchenko]

On Mon, Apr 20, 2026 at 11:08:01AM +0000, Joshua Crofts wrote:
> Clean up headers by removing proxy kernel.h header and adding new
> headers to ensure atomicity (array_size.h, dev_printk.h, types.h,
> asm/byteorder.h). Removed unused headers (slab.h, iio/sysfs.h,
> iio/trigger.h) and added minmax.h, property.h, wait.h, irqreturn.h to
> mitigate transient dependencies during compilation.
> 
> Audited using the include-what-you-use tool.

Seems this implied sorting as well. As Jonathan says somewhere else
he prefers to split sorting and IWYU patches.

...

>  #include <linux/iio/iio.h>
> -#include <linux/iio/sysfs.h>
>  #include <linux/iio/buffer.h>
> -#include <linux/iio/trigger.h>
>  #include <linux/iio/trigger_consumer.h>
>  #include <linux/iio/triggered_buffer.h>

Also note that this group left unsorted.

>  
> -- 
> 2.47.3
> 

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH 1/4] iio: magnetometer: ak8975: header cleanup

[Joshua Crofts]

On Mon, 20 Apr 2026 at 19:18, Andy Shevchenko
<andriy.shevchenko@intel.com> wrote:
> Seems this implied sorting as well. As Jonathan says somewhere else
> he prefers to split sorting and IWYU patches.

Okay, I'll return it pre-sort.

-- 
Kind regards

CJD

Re: [PATCH 1/4] iio: magnetometer: ak8975: header cleanup

[Jonathan Cameron]

On Tue, 21 Apr 2026 08:16:42 +0200
Joshua Crofts <joshua.crofts1@gmail.com> wrote:

> On Mon, 20 Apr 2026 at 19:18, Andy Shevchenko
> <andriy.shevchenko@intel.com> wrote:
> > Seems this implied sorting as well. As Jonathan says somewhere else
> > he prefers to split sorting and IWYU patches.  
> 
> Okay, I'll return it pre-sort.
> 
What I'd prefer is sort, then changes.

That way it's easy to see what is removed or added rather than moved.

Thanks,

Jonathan

[PATCH 2/4] iio: magnetometer: ak8975: replace usleep_range() with fsleep()

[Joshua Crofts]

Replace usleep_range() calls with fsleep(), passing the minimum value
required by the sensor for hardware delays.

fsleep() automatically selects the optimal sleep mechanism, simplifying
driver code and time management.

Signed-off-by: Joshua Crofts <joshua.crofts1@gmail.com>
---
 drivers/iio/magnetometer/ak8975.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/iio/magnetometer/ak8975.c b/drivers/iio/magnetometer/ak8975.c
index f45dca0a6f0..d4ba96a5630 100644
--- a/drivers/iio/magnetometer/ak8975.c
+++ b/drivers/iio/magnetometer/ak8975.c
@@ -461,7 +461,7 @@ static int ak8975_power_on(const struct ak8975_data *data)
 	 * and the minimum wait time before mode setting is 100us, in
 	 * total 300us. Add some margin and say minimum 500us here.
 	 */
-	usleep_range(500, 1000);
+	fsleep(500);
 	return 0;
 }
 
@@ -551,7 +551,7 @@ static int ak8975_set_mode(struct ak8975_data *data, enum ak_ctrl_mode mode)
 
 	data->cntl_cache = regval;
 	/* After mode change wait at least 100us */
-	usleep_range(100, 500);
+	fsleep(100);
 
 	return 0;
 }
-- 
2.47.3

Re: [PATCH 2/4] iio: magnetometer: ak8975: replace usleep_range() with fsleep()

[Andy Shevchenko]

On Mon, Apr 20, 2026 at 11:08:02AM +0000, Joshua Crofts wrote:
> Replace usleep_range() calls with fsleep(), passing the minimum value
> required by the sensor for hardware delays.
> 
> fsleep() automatically selects the optimal sleep mechanism, simplifying
> driver code and time management.

...

> -	usleep_range(500, 1000);
> +	fsleep(500);

Seems code is inconsistent in usage of blank lines before return statements.
I would add one while at it.

>  	return 0;

...

>  	/* After mode change wait at least 100us */
> -	usleep_range(100, 500);
> +	fsleep(100);
>  
>  	return 0;

(Left for the context for the above.)

-- 
With Best Regards,
Andy Shevchenko

[PATCH 3/4] iio: magnetometer: ak8975: change 'u8*' to 'u8 *' in cast

[Joshua Crofts]

Change 'u8*' cast to 'u8 *' as the former triggers a checkpatch error.

Signed-off-by: Joshua Crofts <joshua.crofts1@gmail.com>
---
 drivers/iio/magnetometer/ak8975.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/iio/magnetometer/ak8975.c b/drivers/iio/magnetometer/ak8975.c
index d4ba96a5630..c57ff6d99f3 100644
--- a/drivers/iio/magnetometer/ak8975.c
+++ b/drivers/iio/magnetometer/ak8975.c
@@ -760,7 +760,7 @@ static int ak8975_read_axis(struct iio_dev *indio_dev, int index, int *val)
 
 	ret = i2c_smbus_read_i2c_block_data_or_emulated(
 			client, def->data_regs[index],
-			sizeof(rval), (u8*)&rval);
+			sizeof(rval), (u8 *)&rval);
 	if (ret < 0)
 		goto exit;
 
-- 
2.47.3

Re: [PATCH 3/4] iio: magnetometer: ak8975: change 'u8*' to 'u8 *' in cast

[Andy Shevchenko]

On Mon, Apr 20, 2026 at 11:08:03AM +0000, Joshua Crofts wrote:
> Change 'u8*' cast to 'u8 *' as the former triggers a checkpatch error.

...

>  	ret = i2c_smbus_read_i2c_block_data_or_emulated(
>  			client, def->data_regs[index],
> -			sizeof(rval), (u8*)&rval);
> +			sizeof(rval), (u8 *)&rval);

Strange no warnings on the open-ended function call on the first line
(trailing "("). That said, I would rather see the result as

	ret = i2c_smbus_read_i2c_block_data_or_emulated(client,
							def->data_regs[index],
							sizeof(rval),
							(u8 *)&rval);


-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH 3/4] iio: magnetometer: ak8975: change 'u8*' to 'u8 *' in cast

[Joshua Crofts]

On Mon, 20 Apr 2026 at 19:23, Andy Shevchenko
<andriy.shevchenko@intel.com> wrote:
> Strange no warnings on the open-ended function call on the first line
> (trailing "("). That said, I would rather see the result as
>
>         ret = i2c_smbus_read_i2c_block_data_or_emulated(client,
>                                                         def->data_regs[index],
>                                                         sizeof(rval),
>                                                         (u8 *)&rval);

Checkpatch seems to miss quite a few issues. Will fix.

-- 
Kind regards

CJD

[PATCH 4/4] iio: magnetometer: ak8975: modernize polling loops with iopoll() macros

[Joshua Crofts]

The driver currently uses while loops and msleep() for polling during
conversion waits.

Replace the custom polling loops with readx_poll_timeout() and
read_poll_timeout() macros from <linux/iopoll.h>. This reduces
boilerplate, standardizes timeout handling and improves overall code
readability, keeping the original timing and error behaviour.

Assisted-by: Gemini:3.1-Pro
Signed-off-by: Joshua Crofts <joshua.crofts1@gmail.com>
---
 drivers/iio/magnetometer/ak8975.c | 44 ++++++++++++++-----------------
 1 file changed, 20 insertions(+), 24 deletions(-)

diff --git a/drivers/iio/magnetometer/ak8975.c b/drivers/iio/magnetometer/ak8975.c
index c57ff6d99f3..86b4149e77b 100644
--- a/drivers/iio/magnetometer/ak8975.c
+++ b/drivers/iio/magnetometer/ak8975.c
@@ -15,6 +15,7 @@
 #include <linux/gpio/consumer.h>
 #include <linux/i2c.h>
 #include <linux/interrupt.h>
+#include <linux/iopoll.h>
 #include <linux/irqreturn.h>
 #include <linux/minmax.h>
 #include <linux/mod_devicetable.h>
@@ -651,17 +652,15 @@ static int ak8975_setup(struct i2c_client *client)
 static int wait_conversion_complete_gpio(struct ak8975_data *data)
 {
 	struct i2c_client *client = data->client;
-	u32 timeout_ms = AK8975_MAX_CONVERSION_TIMEOUT;
 	int ret;
+	int val;
 
 	/* Wait for the conversion to complete. */
-	while (timeout_ms) {
-		msleep(AK8975_CONVERSION_DONE_POLL_TIME);
-		if (gpiod_get_value(data->eoc_gpiod))
-			break;
-		timeout_ms -= AK8975_CONVERSION_DONE_POLL_TIME;
-	}
-	if (!timeout_ms) {
+	ret = readx_poll_timeout(gpiod_get_value, data->eoc_gpiod, val, val != 0,
+							 AK8975_CONVERSION_DONE_POLL_TIME * 1000,
+							 AK8975_MAX_CONVERSION_TIMEOUT * 1000);
+
+	if (ret) {
 		dev_err(&client->dev, "Conversion timeout happened\n");
 		return -EINVAL;
 	}
@@ -676,30 +675,27 @@ static int wait_conversion_complete_gpio(struct ak8975_data *data)
 static int wait_conversion_complete_polled(struct ak8975_data *data)
 {
 	struct i2c_client *client = data->client;
-	u8 read_status;
-	u32 timeout_ms = AK8975_MAX_CONVERSION_TIMEOUT;
 	int ret;
+	int val;
 
 	/* Wait for the conversion to complete. */
-	while (timeout_ms) {
-		msleep(AK8975_CONVERSION_DONE_POLL_TIME);
-		ret = i2c_smbus_read_byte_data(client,
-					       data->def->ctrl_regs[ST1]);
-		if (ret < 0) {
-			dev_err(&client->dev, "Error in reading ST1\n");
-			return ret;
-		}
-		read_status = ret;
-		if (read_status)
-			break;
-		timeout_ms -= AK8975_CONVERSION_DONE_POLL_TIME;
+	ret = read_poll_timeout(i2c_smbus_read_byte_data, val, val != 0,
+							AK8975_CONVERSION_DONE_POLL_TIME * 1000,
+							AK8975_MAX_CONVERSION_TIMEOUT * 1000,
+							true,
+							client, data->def->ctrl_regs[ST1]);
+
+	if (val < 0) {
+		dev_err(&client->dev, "Error in reading ST1\n");
+		return val;
 	}
-	if (!timeout_ms) {
+
+	if (ret == -ETIMEDOUT) {
 		dev_err(&client->dev, "Conversion timeout happened\n");
 		return -EINVAL;
 	}
 
-	return read_status;
+	return val;
 }
 
 /* Returns 0 if the end of conversion interrupt occurred or -ETIME otherwise */
-- 
2.47.3

Re: [PATCH 4/4] iio: magnetometer: ak8975: modernize polling loops with iopoll() macros

[Andy Shevchenko]

On Mon, Apr 20, 2026 at 11:08:04AM +0000, Joshua Crofts wrote:
> The driver currently uses while loops and msleep() for polling during
> conversion waits.
> 
> Replace the custom polling loops with readx_poll_timeout() and
> read_poll_timeout() macros from <linux/iopoll.h>. This reduces
> boilerplate, standardizes timeout handling and improves overall code
> readability, keeping the original timing and error behaviour.

...

>  	/* Wait for the conversion to complete. */
> -	while (timeout_ms) {
> -		msleep(AK8975_CONVERSION_DONE_POLL_TIME);
> -		if (gpiod_get_value(data->eoc_gpiod))
> -			break;
> -		timeout_ms -= AK8975_CONVERSION_DONE_POLL_TIME;
> -	}
> -	if (!timeout_ms) {
> +	ret = readx_poll_timeout(gpiod_get_value, data->eoc_gpiod, val, val != 0,

' != 0' is not needed, but perhaps okay in this case to explicitly hint that
this is used as boolean conditional.

> +							 AK8975_CONVERSION_DONE_POLL_TIME * 1000,
> +							 AK8975_MAX_CONVERSION_TIMEOUT * 1000);

Somebody broke the indentation...

> +

...and added unneeded blank line.

Also, use multipliers from time.h, id est USEC_PER_MSEC.

> +	if (ret) {
>  		dev_err(&client->dev, "Conversion timeout happened\n");
>  		return -EINVAL;
>  	}

...

>  static int wait_conversion_complete_polled(struct ak8975_data *data)

Same comments as per above.

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH 4/4] iio: magnetometer: ak8975: modernize polling loops with iopoll() macros

[Joshua Crofts]

On Mon, 20 Apr 2026 at 19:26, Andy Shevchenko
<andriy.shevchenko@intel.com> wrote:
> ' != 0' is not needed, but perhaps okay in this case to explicitly hint that
> this is used as boolean conditional.

Yeah, that was my intention, I thought that (...val, val, ...) was a
bit confusing.

> Somebody broke the indentation...
>
> > +
>
> ...and added unneeded blank line.

I should probably just stick to one editor.

> Also, use multipliers from time.h, id est USEC_PER_MSEC.

Will do.

-- 
Kind regards

CJD