Skip to contentSkip to conversation

iio: magnetometer: ak8975: modernize and cleanup driver

PATCH Applied Joshua Crofts 5 participants 75 messages May 5 – May 7 IIO

lore
Share & tools
lore
#0 lore raw

v5 cover iio: magnetometer: ak8975: modernize and cleanup driver

This series modernizes and cleans up the ak8975 driver, migrating to more modern kernel APIs and cleaning up legacy code. Patches 7 through 17 were provided by Andy Shevchenko.

The series was compile-tested and ran in QEMU with i2c-stub and no issues were found.

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
- switching driver to use devm_* functions

---
Changes in v2:
- PATCH 1: reverted alphabetical header ordering per logical change separation
- PATCH 2: added newline between fsleep() and return
- PATCH 3: fixed open parenthesis formatting
- PATCH 4: fixed whitespace issues, added <linux/time.h> for USEC_PER_MSEC macro

Changes in v3:
- PATCH 1 & 2: split alphabetical ordering and header removal into two separate patches
- PATCH 4: commit message fix
- PATCH 5 & 6: split error handling and polling loop replacement into two separate patches

Changes in v4:
- PATCH 7-17: am Andy's changes into this series
- PATCH 2, 5, 6: fixup Andy's changes to existing patches

Changes in v5:
- PATCH 6 & 7: reorder patch series, edit commit messages
- PATCH 8: add check for negative val from readx_*() call
- Various cleanups

---

---
Andy Shevchenko (10):
iio: magnetometer: ak8975: avoid using temporary variable
iio: magnetometer: ak8975: drop duplicate NULL check
iio: magnetometer: ak8975: remove duplicate error message
iio: magnetometer: ak8975: reduce usage of magic lengths of the buffer
iio: magnetometer: ak8975: unify return code variable name
iio: magnetometer: ak8975: switch to using managed resources
iio: magnetometer: ak8975: consistently use 'data' parameter
iio: magnetometer: ak8975: unify messages with help of dev_err_probe()
iio: magnetometer: ak8975: use temporary variable for struct device
iio: magnetometer: ak8975: make use of the macros from bits.h

Joshua Crofts (8):
iio: magnetometer: ak8975: sort headers alphabetically
iio: magnetometer: ak8975: update headers per IWYU principle
iio: magnetometer: ak8975: replace usleep_range() with fsleep()
iio: magnetometer: ak8975: change 'u8*' to 'u8 *' in cast
iio: magnetometer: ak8975: fix wrong errno on return
iio: magnetometer: ak8975: pass conversion timeouts as arguments
iio: magnetometer: ak8975: modernize polling loops with iopoll() macros
iio: magnetometer: ak8975: check if gpiod read was successful

 drivers/iio/magnetometer/ak8975.c | 369 +++++++++++++++++++-------------------
 1 file changed, 180 insertions(+), 189 deletions(-)

---
base-commit: 440c14da1af1cc6cc9bb18841603711d845ed900
change-id: 20260505-magnetometer-fixes-d3a96180ab4c

Best regards,

Signature · 1 line
Joshua Crofts <[email protected]>
#1 lore raw

PATCH v5 1/18 iio: magnetometer: ak8975: sort headers alphabetically

Sort include headers alphabetically to improve coding style and readability.

No functional change.

---
 drivers/iio/magnetometer/ak8975.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)
drivers/iio/magnetometer/ak8975.c +10 -10
index 44782c26698bcac5ebfc62ce3740f361078fd0ce..569cd6fa839aec6ee8758a261d0e659fc20131c6 100644
@@ -7,23 +7,23 @@
  * Copyright (c) 2010, NVIDIA Corporation.
  */
 
-#include <linux/module.h>
-#include <linux/mod_devicetable.h>
-#include <linux/kernel.h>
-#include <linux/slab.h>
+#include <linux/bitops.h>
+#include <linux/delay.h>
+#include <linux/err.h>
+#include <linux/gpio/consumer.h>
 #include <linux/i2c.h>
 #include <linux/interrupt.h>
-#include <linux/err.h>
+#include <linux/kernel.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/regulator/consumer.h>
+#include <linux/slab.h>
 
+#include <linux/iio/buffer.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>
Signature · 1 line
2.47.3
Reviewed-by on v5 1/18
#2 lore raw

Re: [PATCH v5 01/18] iio: magnetometer: ak8975: sort headers alphabetically

Quoted 13 lines · On Tue, May 5, 2026 at 6:46 AM Joshua Crofts via B4 Relay…

From: Joshua Crofts <[email protected]>

Sort include headers alphabetically to improve coding style and readability.

No functional change.

---
 drivers/iio/magnetometer/ak8975.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

[snip]

Quoted 1 line

LGTM.

best regards,
max

maintainer
#3 lore raw

Re: [PATCH v5 01/18] iio: magnetometer: ak8975: sort headers alphabetically

Quoted 18 lines · On Tue, 5 May 2026 18:46:15 -0500 Maxwell Doose <[email protected]> wrote:
Quoted 13 lines · On Tue, May 5, 2026 at 6:46 AM Joshua Crofts via B4 Relay…

From: Joshua Crofts <[email protected]>

Sort include headers alphabetically to improve coding style and readability.

No functional change.

---
 drivers/iio/magnetometer/ak8975.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

[snip]

Quoted 1 line
Quoted 6 lines

LGTM.

best regards,
max

Too many patches floating around so I'm going to pick up the start of this series at least. Applied this one to the testing branch of iio.git.

#4 lore raw

PATCH v5 2/18 iio: magnetometer: ak8975: update headers per IWYU principle

In reply to Joshua Crofts

Remove kernel.h proxy header and unused headers (slab.h, iio/sysfs.h, iio/trigger.h). Add missing headers to ensure atomicity (array_size.h, dev_printk.h, asm/byteorder.h, irqreturn.h, minmax.h, property.h, types.h, wait.h).

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

---
 drivers/iio/magnetometer/ak8975.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)
drivers/iio/magnetometer/ak8975.c +9 -4
index 569cd6fa839aec6ee8758a261d0e659fc20131c6..dac0a90c33eebd43f8aeed846c9cbf5fd1a6ab91 100644
@@ -7,24 +7,29 @@
  * Copyright (c) 2010, NVIDIA Corporation.
  */
 
+#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/kernel.h>
+#include <linux/iopoll.h>
+#include <linux/minmax.h>
 #include <linux/mod_devicetable.h>
 #include <linux/module.h>
 #include <linux/mutex.h>
 #include <linux/pm_runtime.h>
+#include <linux/property.h>
 #include <linux/regulator/consumer.h>
-#include <linux/slab.h>
+#include <linux/types.h>
+#include <linux/wait.h>
+
+#include <asm/byteorder.h>
 
 #include <linux/iio/buffer.h>
 #include <linux/iio/iio.h>
-#include <linux/iio/sysfs.h>
-#include <linux/iio/trigger.h>
 #include <linux/iio/trigger_consumer.h>
 #include <linux/iio/triggered_buffer.h>
 
Signature · 1 line
2.47.3
Reviewed-by on v5 2/18
#5 lore raw

Re: [PATCH v5 02/18] iio: magnetometer: ak8975: update headers per IWYU principle

Quoted 16 lines · On Tue, May 5, 2026 at 6:58 AM Joshua Crofts via B4 Relay…

From: Joshua Crofts <[email protected]>

Remove kernel.h proxy header and unused headers (slab.h, iio/sysfs.h, iio/trigger.h). Add missing headers to ensure atomicity (array_size.h, dev_printk.h, asm/byteorder.h, irqreturn.h, minmax.h, property.h, types.h, wait.h).

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

---
 drivers/iio/magnetometer/ak8975.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

[snip]

Quoted 1 line

LGTM.

best regards,
max

maintainer
#6 lore raw

Re: [PATCH v5 02/18] iio: magnetometer: ak8975: update headers per IWYU principle

Quoted 18 lines · On Tue, 5 May 2026 19:38:55 -0500 Maxwell Doose <[email protected]> wrote:
Quoted 8 lines · On Tue, May 5, 2026 at 6:58 AM Joshua Crofts via B4 Relay…

From: Joshua Crofts <[email protected]>

Remove kernel.h proxy header and unused headers (slab.h, iio/sysfs.h, iio/trigger.h). Add missing headers to ensure atomicity (array_size.h, dev_printk.h, asm/byteorder.h, irqreturn.h, minmax.h, property.h, types.h, wait.h).

Quoted 8 lines

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

---
 drivers/iio/magnetometer/ak8975.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)
Quoted 6 lines

[snip]

Quoted 1 line

LGTM.

Applied to the testing branch of iio.git.
If you do a v6 base either base on that (and state it) or just drop the ones I've applied and don't state a base.

Thanks,

Jonathan

Quoted 3 lines

best regards,
max

#7 lore raw

PATCH v5 3/18 iio: magnetometer: ak8975: replace usleep_range() with fsleep()

In reply to 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.

---
 drivers/iio/magnetometer/ak8975.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/iio/magnetometer/ak8975.c +3 -2
index dac0a90c33eebd43f8aeed846c9cbf5fd1a6ab91..a84a05a1b64280e26c428e9752a39380e386c234 100644
@@ -461,7 +461,8 @@ 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 +552,7 @@ 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;
 }
Signature · 1 line
2.47.3
#8 lore raw

Re: [PATCH v5 03/18] iio: magnetometer: ak8975: replace usleep_range() with fsleep()

Hi Joshua,

Quoted 6 lines · On Tue, May 5, 2026 at 6:53 AM Joshua Crofts via B4 Relay…

From: Joshua Crofts <[email protected]>

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

Quoted 7 lines

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

---
 drivers/iio/magnetometer/ak8975.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

[snip]

Looks fine overall but has this been tested on actual hardware? Just want to double check since this touches the set_mode and power_on logic. Also I will say it may be better to keep usleep_range() since it's more explicit, which will likely be better for the future to make it obvious. Also, I was taking a look at the docs for fsleep() and it provides "maximum 25% slack", which means we're stripping away almost 375us on one of these calls. And besides, it would eventually just call usleep_range() anyways.

best regards,
max

#9 lore raw

Re: [PATCH v5 03/18] iio: magnetometer: ak8975: replace usleep_range() with fsleep()

Quoted 11 lines · Maxwell Doose

Hi Joshua,

Quoted 6 lines · On Tue, May 5, 2026 at 6:53 AM Joshua Crofts via B4 Relay…

From: Joshua Crofts <[email protected]>

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

Quoted 8 lines
Quoted 7 lines

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

---
 drivers/iio/magnetometer/ak8975.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

[snip]

Hi Max,

Quoted 3 lines

Looks fine overall but has this been tested on actual hardware? Just want to double check since this touches the set_mode and power_on logic.

Fair enough, this series mostly concerned code cleanup and moving away from older driver programming practices, so it shouldn't mess with the functioning logic as much - however it would be more than welcome if someone had the hardware to test it.

Quoted 6 lines

Also I will say it may be better to keep usleep_range() since it's more explicit, which will likely be better for the future to make it obvious. Also, I was taking a look at the docs for fsleep() and it provides "maximum 25% slack", which means we're stripping away almost 375us on one of these calls. And besides, it would eventually just call usleep_range() anyways.

I disagree here. If you look at the documentation for msleep(), you can see that "slack" is actually an addition to the delay time, not a subtraction. fsleep()
will always guarantee a delay of at least what's supplied as the parameter.

About fsleep() translating to usleep_range() - yes, you're correct about that, however not only it's more modern practice, it also doesn't require the programmer to add an additional arbitrary max value, fsleep() just adds it itself.
As for explicitness, I think fsleep(500) is explicit enough, as described above.

Signature · 3 lines
Kind regards

CJD
#10 lore raw

Re: [PATCH v5 03/18] iio: magnetometer: ak8975: replace usleep_range() with fsleep()

Quoted 19 lines · Joshua Crofts

Hi Max,

Quoted 3 lines

Looks fine overall but has this been tested on actual hardware? Just want to double check since this touches the set_mode and power_on logic.

Fair enough, this series mostly concerned code cleanup and moving away from older driver programming practices, so it shouldn't mess with the functioning logic as much - however it would be more than welcome if someone had the hardware to test it.

Quoted 6 lines

Also I will say it may be better to keep usleep_range() since it's more explicit, which will likely be better for the future to make it obvious. Also, I was taking a look at the docs for fsleep() and it provides "maximum 25% slack", which means we're stripping away almost 375us on one of these calls. And besides, it would eventually just call usleep_range() anyways.

Quoted 5 lines

I disagree here. If you look at the documentation for msleep(), you can see that "slack" is actually an addition to the delay time, not a subtraction. fsleep()
will always guarantee a delay of at least what's supplied as the parameter.

Sorry if there was any confusion, I just double checked the docs myself, I was saying if we do fsleep(100); wouldn't that mean the allowable range is 100-125us? Might put some extra strain vs. the 400us we had before. If I'm incorrect on that you can call me out on that.

Quoted 7 lines

About fsleep() translating to usleep_range() - yes, you're correct about that, however not only it's more modern practice, it also doesn't require the programmer to add an additional arbitrary max value, fsleep() just adds it itself.
As for explicitness, I think fsleep(500) is explicit enough, as described above.

Probably true, but like above my only concern is timing.

best regards,
max

Quoted 5 lines

--
Kind regards

CJD

#11 lore raw

Re: [PATCH v5 03/18] iio: magnetometer: ak8975: replace usleep_range() with fsleep()

Quoted 27 lines · Maxwell Doose
Quoted 19 lines · Joshua Crofts

Hi Max,

Quoted 3 lines

Looks fine overall but has this been tested on actual hardware? Just want to double check since this touches the set_mode and power_on logic.

Fair enough, this series mostly concerned code cleanup and moving away from older driver programming practices, so it shouldn't mess with the functioning logic as much - however it would be more than welcome if someone had the hardware to test it.

Quoted 6 lines

Also I will say it may be better to keep usleep_range() since it's more explicit, which will likely be better for the future to make it obvious. Also, I was taking a look at the docs for fsleep() and it provides "maximum 25% slack", which means we're stripping away almost 375us on one of these calls. And besides, it would eventually just call usleep_range() anyways.

Quoted 5 lines

I disagree here. If you look at the documentation for msleep(), you can see that "slack" is actually an addition to the delay time, not a subtraction. fsleep()
will always guarantee a delay of at least what's supplied as the parameter.

Quoted 5 lines

Sorry if there was any confusion, I just double checked the docs myself, I was saying if we do fsleep(100); wouldn't that mean the allowable range is 100-125us? Might put some extra strain vs. the 400us we had before. If I'm incorrect on that you can call me out on that.

Okay, now I understand. From a scheduling perspective you're right, it would be less strain. At the same time I don't think that was the original intention - the author probably just slipped a random value relatively in range to satisfy the max parameter.

Either way, we should perhaps take this up with the reviewers, as Jonathan and Andy (and Sashiko) didn't have any issue with this.

Signature · 3 lines
Kind regards

CJD
#12 lore raw

Re: [PATCH v5 03/18] iio: magnetometer: ak8975: replace usleep_range() with fsleep()

Quoted 9 lines · Joshua Crofts

Okay, now I understand. From a scheduling perspective you're right, it would be less strain. At the same time I don't think that was the original intention - the author probably just slipped a random value relatively in range to satisfy the max parameter.

Either way, we should perhaps take this up with the reviewers, as Jonathan and Andy (and Sashiko) didn't have any issue with this.

Sounds like a good idea, they'll likely have a better idea of this driver than I do.

best regards,
max

Quoted 4 lines

--
Kind regards

CJD

#13 lore raw

Re: [PATCH v5 03/18] iio: magnetometer: ak8975: replace usleep_range() with fsleep()

In reply to Joshua Crofts
Quoted 9 lines · Joshua Crofts
Quoted 8 lines · Maxwell Doose
Quoted 6 lines · On Tue, May 5, 2026 at 6:53 AM Joshua Crofts via B4 Relay…

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.

[snip]

Quoted 8 lines
Quoted 3 lines

Looks fine overall but has this been tested on actual hardware? Just want to double check since this touches the set_mode and power_on logic.

Fair enough, this series mostly concerned code cleanup and moving away from older driver programming practices, so it shouldn't mess with the functioning logic as much - however it would be more than welcome if someone had the hardware to test it.

This simple patch doesn't change the minimum time to sleep, assuming the driver was tested previously (and it had to be), the given change is very unlikely to introduce regressions. Also note that the similar changes have been done all around the kernel in the past few month and I haven't heard it brought a regression to the users (but if you, Maxwell, have some pointers to otherwise, please share, I would really like to see such a report!).

Quoted 12 lines
Quoted 6 lines

Also I will say it may be better to keep usleep_range() since it's more explicit, which will likely be better for the future to make it obvious. Also, I was taking a look at the docs for fsleep() and it provides "maximum 25% slack", which means we're stripping away almost 375us on one of these calls. And besides, it would eventually just call usleep_range() anyways.

I disagree here. If you look at the documentation for msleep(), you can see that "slack" is actually an addition to the delay time, not a subtraction. fsleep()
will always guarantee a delay of at least what's supplied as the parameter.

Quoted 5 lines

About fsleep() translating to usleep_range() - yes, you're correct about that, however not only it's more modern practice, it also doesn't require the programmer to add an additional arbitrary max value, fsleep() just adds it itself.
As for explicitness, I think fsleep(500) is explicit enough, as described above.

Yes, fsleep() is recommended way of non-atomic sleeps with a microsecond granularity.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#14 lore raw

Re: [PATCH v5 03/18] iio: magnetometer: ak8975: replace usleep_range() with fsleep()

Quoted 1 line · Andy Shevchenko

[snip]

Quoted 8 lines

This simple patch doesn't change the minimum time to sleep, assuming the driver was tested previously (and it had to be), the given change is very unlikely to introduce regressions. Also note that the similar changes have been done all around the kernel in the past few month and I haven't heard it brought a regression to the users (but if you, Maxwell, have some pointers to otherwise, please share, I would really like to see such a report!).

Alright, fair enough. I'm just not as familiar as I'd like to be with the fsleep() stuff and wanted to make sure the timings were good before guaranteeing anything.

best regards,
max

Quoted 9 lines

Yes, fsleep() is recommended way of non-atomic sleeps with a microsecond granularity.

--
With Best Regards,
Andy Shevchenko

maintainer
#15 lore raw

Re: [PATCH v5 03/18] iio: magnetometer: ak8975: replace usleep_range() with fsleep()

Quoted 13 lines · On Wed, 6 May 2026 01:35:32 -0500 Maxwell Doose <[email protected]> wrote:
Quoted 1 line · Andy Shevchenko

[snip]

Quoted 8 lines

This simple patch doesn't change the minimum time to sleep, assuming the driver was tested previously (and it had to be), the given change is very unlikely to introduce regressions. Also note that the similar changes have been done all around the kernel in the past few month and I haven't heard it brought a regression to the users (but if you, Maxwell, have some pointers to otherwise, please share, I would really like to see such a report!).

Quoted 6 lines

Alright, fair enough. I'm just not as familiar as I'd like to be with the fsleep() stuff and wanted to make sure the timings were good before guaranteeing anything.

best regards,
max

Applied this one to the testing branch of iio.git

Thanks,

Jonathan

Quoted 12 lines
Quoted 12 lines

Yes, fsleep() is recommended way of non-atomic sleeps with a microsecond granularity.

--
With Best Regards,
Andy Shevchenko

#16 lore raw

PATCH v5 4/18 iio: magnetometer: ak8975: change 'u8*' to 'u8 *' in cast

In reply to Joshua Crofts

Change 'u8*' cast to 'u8 *' as the former triggers a checkpatch error. Also fix the indentation of parameters in i2c_smbus_read_i2c_block_data_or_emulated() function.

No functional change.

---
 drivers/iio/magnetometer/ak8975.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/iio/magnetometer/ak8975.c +4 -3
index a84a05a1b64280e26c428e9752a39380e386c234..6ca8c24b9fba452b6fed8155957a8e5fcbc6511c 100644
@@ -759,9 +759,10 @@ static int ak8975_read_axis(struct iio_dev *indio_dev, int index, int *val)
 	if (ret)
 		goto exit;
 
-	ret = i2c_smbus_read_i2c_block_data_or_emulated(
-			client, def->data_regs[index],
-			sizeof(rval), (u8*)&rval);
+	ret = i2c_smbus_read_i2c_block_data_or_emulated(client,
+							def->data_regs[index],
+							sizeof(rval),
+							(u8 *)&rval);
 	if (ret < 0)
 		goto exit;
 
Signature · 1 line
2.47.3
Reviewed-by on v5 4/18
#17 lore raw

Re: [PATCH v5 04/18] iio: magnetometer: ak8975: change 'u8*' to 'u8 *' in cast

Quoted 14 lines · On Tue, May 5, 2026 at 6:46 AM Joshua Crofts via B4 Relay…

From: Joshua Crofts <[email protected]>

Change 'u8*' cast to 'u8 *' as the former triggers a checkpatch error. Also fix the indentation of parameters in i2c_smbus_read_i2c_block_data_or_emulated() function.

No functional change.

---
 drivers/iio/magnetometer/ak8975.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

[snip]

Quoted 1 line

LGTM.

best regards,
max

Reviewed-by on v5 4/18
#18 lore raw

Re: [PATCH v5 04/18] iio: magnetometer: ak8975: change 'u8*' to 'u8 *' in cast

In reply to Joshua Crofts
Quoted 5 lines · On Tue, May 05, 2026 at 01:46:00PM +0200, Joshua Crofts via B4 Relay wrote:

Change 'u8*' cast to 'u8 *' as the former triggers a checkpatch error. Also fix the indentation of parameters in i2c_smbus_read_i2c_block_data_or_emulated() function.

No functional change.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
maintainer Applied
#19 lore raw

Re: [PATCH v5 04/18] iio: magnetometer: ak8975: change 'u8*' to 'u8 *' in cast

Quoted 10 lines · On Wed, 6 May 2026 09:30:54 +0300 Andy Shevchenko…
Quoted 5 lines · On Tue, May 05, 2026 at 01:46:00PM +0200, Joshua Crofts via B4 Relay wrote:

Change 'u8*' cast to 'u8 *' as the former triggers a checkpatch error. Also fix the indentation of parameters in i2c_smbus_read_i2c_block_data_or_emulated() function.

No functional change.

Applied.

#20 lore raw

PATCH v5 5/18 iio: magnetometer: ak8975: fix wrong errno on return

In reply to Joshua Crofts

The driver currently returns -EINVAL on polling timeout instead of -ETIMEDOUT.

Replace return code for -ETIMEDOUT and remove unnecessary error message as -ETIMEDOUT is a standard POSIX error. Also replace instances of -EINVAL in comments.

---
 drivers/iio/magnetometer/ak8975.c | 16 ++++++----------
 1 file changed, 6 insertions(+), 10 deletions(-)
drivers/iio/magnetometer/ak8975.c +6 -10
index 6ca8c24b9fba452b6fed8155957a8e5fcbc6511c..57f50c09cca539c3733f516a1617375e9134c349 100644
@@ -662,10 +662,8 @@ static int wait_conversion_complete_gpio(struct ak8975_data *data)
 			break;
 		timeout_ms -= AK8975_CONVERSION_DONE_POLL_TIME;
 	}
-	if (!timeout_ms) {
-		dev_err(&client->dev, "Conversion timeout happened\n");
-		return -EINVAL;
-	}
+	if (!timeout_ms)
+		return -ETIMEDOUT;
 
 	ret = i2c_smbus_read_byte_data(client, data->def->ctrl_regs[ST1]);
 	if (ret < 0)
@@ -695,15 +693,13 @@ static int wait_conversion_complete_polled(struct ak8975_data *data)
 			break;
 		timeout_ms -= AK8975_CONVERSION_DONE_POLL_TIME;
 	}
-	if (!timeout_ms) {
-		dev_err(&client->dev, "Conversion timeout happened\n");
-		return -EINVAL;
-	}
+	if (!timeout_ms)
+		return -ETIMEDOUT;
 
 	return read_status;
 }
 
-/* Returns 0 if the end of conversion interrupt occurred or -ETIME otherwise */
+/* Returns 0 if the end of conversion interrupt occurred or -ETIMEDOUT otherwise */
 static int wait_conversion_complete_interrupt(struct ak8975_data *data)
 {
 	int ret;
@@ -713,7 +709,7 @@ static int wait_conversion_complete_interrupt(struct ak8975_data *data)
 				 AK8975_DATA_READY_TIMEOUT);
 	clear_bit(0, &data->flags);
 
-	return ret > 0 ? 0 : -ETIME;
+	return ret > 0 ? 0 : -ETIMEDOUT;
 }
 
 static int ak8975_start_read_axis(struct ak8975_data *data,
Signature · 1 line
2.47.3
maintainer Applied
#21 lore raw

Re: [PATCH v5 05/18] iio: magnetometer: ak8975: fix wrong errno on return

Quoted 5 lines · On Tue, 05 May 2026 13:46:01 +0200 Joshua Crofts via B4 Relay…

From: Joshua Crofts <[email protected]>

The driver currently returns -EINVAL on polling timeout instead of -ETIMEDOUT.

Quoted 7 lines

Replace return code for -ETIMEDOUT and remove unnecessary error message as -ETIMEDOUT is a standard POSIX error. Also replace instances of -EINVAL in comments.

Applied.

Tah

#22 lore raw

PATCH v5 6/18 iio: magnetometer: ak8975: pass conversion timeouts as arguments

In reply to Joshua Crofts

Refactor wait_conversion_complete*_() helper function to accept poll and timeout values directly as parameters. This improves the readability of the code and does not rely on hardcoded macros.

Besides that, fix the home grown and obviously wrong in some cases the jiffy-based timeout.

---
 drivers/iio/magnetometer/ak8975.c | 42 +++++++++++++++++++--------------------
 1 file changed, 20 insertions(+), 22 deletions(-)
drivers/iio/magnetometer/ak8975.c +20 -22
index 57f50c09cca539c3733f516a1617375e9134c349..83878ea28b91d8316d2cc31e57fde6da957af689 100644
@@ -16,6 +16,7 @@
 #include <linux/i2c.h>
 #include <linux/interrupt.h>
 #include <linux/iopoll.h>
+#include <linux/jiffies.h>
 #include <linux/minmax.h>
 #include <linux/mod_devicetable.h>
 #include <linux/module.h>
@@ -132,13 +133,6 @@
 
 #define AK09912_MAX_REGS		AK09912_REG_ASAZ
 
-/*
- * Miscellaneous values.
- */
-#define AK8975_MAX_CONVERSION_TIMEOUT	500
-#define AK8975_CONVERSION_DONE_POLL_TIME 10
-#define AK8975_DATA_READY_TIMEOUT	((100*HZ)/1000)
-
 /*
  * Precalculate scale factor (in Gauss units) for each axis and
  * store in the device data.
@@ -649,18 +643,19 @@ static int ak8975_setup(struct i2c_client *client)
 	return 0;
 }
 
-static int wait_conversion_complete_gpio(struct ak8975_data *data)
+static int wait_conversion_complete_gpio(struct ak8975_data *data,
+					 unsigned int poll_ms,
+					 unsigned int timeout_ms)
 {
 	struct i2c_client *client = data->client;
-	u32 timeout_ms = AK8975_MAX_CONVERSION_TIMEOUT;
 	int ret;
 
 	/* Wait for the conversion to complete. */
 	while (timeout_ms) {
-		msleep(AK8975_CONVERSION_DONE_POLL_TIME);
+		msleep(poll_ms);
 		if (gpiod_get_value(data->eoc_gpiod))
 			break;
-		timeout_ms -= AK8975_CONVERSION_DONE_POLL_TIME;
+		timeout_ms -= poll_ms;
 	}
 	if (!timeout_ms)
 		return -ETIMEDOUT;
@@ -672,16 +667,17 @@ static int wait_conversion_complete_gpio(struct ak8975_data *data)
 	return ret;
 }
 
-static int wait_conversion_complete_polled(struct ak8975_data *data)
+static int wait_conversion_complete_polled(struct ak8975_data *data,
+					   unsigned int poll_ms,
+					   unsigned int timeout_ms)
 {
 	struct i2c_client *client = data->client;
 	u8 read_status;
-	u32 timeout_ms = AK8975_MAX_CONVERSION_TIMEOUT;
 	int ret;
 
 	/* Wait for the conversion to complete. */
 	while (timeout_ms) {
-		msleep(AK8975_CONVERSION_DONE_POLL_TIME);
+		msleep(poll_ms);
 		ret = i2c_smbus_read_byte_data(client,
 					       data->def->ctrl_regs[ST1]);
 		if (ret < 0) {
@@ -691,7 +687,7 @@ static int wait_conversion_complete_polled(struct ak8975_data *data)
 		read_status = ret;
 		if (read_status)
 			break;
-		timeout_ms -= AK8975_CONVERSION_DONE_POLL_TIME;
+		timeout_ms -= poll_ms;
 	}
 	if (!timeout_ms)
 		return -ETIMEDOUT;
@@ -700,13 +696,14 @@ static int wait_conversion_complete_polled(struct ak8975_data *data)
 }
 
 /* Returns 0 if the end of conversion interrupt occurred or -ETIMEDOUT otherwise */
-static int wait_conversion_complete_interrupt(struct ak8975_data *data)
+static int wait_conversion_complete_interrupt(struct ak8975_data *data,
+					      unsigned int timeout_ms)
 {
 	int ret;
 
 	ret = wait_event_timeout(data->data_ready_queue,
 				 test_bit(0, &data->flags),
-				 AK8975_DATA_READY_TIMEOUT);
+				 msecs_to_jiffies(timeout_ms));
 	clear_bit(0, &data->flags);
 
 	return ret > 0 ? 0 : -ETIMEDOUT;
@@ -715,9 +712,10 @@ static int wait_conversion_complete_interrupt(struct ak8975_data *data)
 static int ak8975_start_read_axis(struct ak8975_data *data,
 				  const struct i2c_client *client)
 {
-	/* Set up the device for taking a sample. */
-	int ret = ak8975_set_mode(data, MODE_ONCE);
+	int ret;
 
+	/* Set up the device for taking a sample. */
+	ret = ak8975_set_mode(data, MODE_ONCE);
 	if (ret < 0) {
 		dev_err(&client->dev, "Error in setting operating mode\n");
 		return ret;
@@ -725,11 +723,11 @@ static int ak8975_start_read_axis(struct ak8975_data *data,
 
 	/* Wait for the conversion to complete. */
 	if (data->eoc_irq)
-		ret = wait_conversion_complete_interrupt(data);
+		ret = wait_conversion_complete_interrupt(data, 100);
 	else if (data->eoc_gpiod)
-		ret = wait_conversion_complete_gpio(data);
+		ret = wait_conversion_complete_gpio(data, 10, 500);
 	else
-		ret = wait_conversion_complete_polled(data);
+		ret = wait_conversion_complete_polled(data, 10, 500);
 	if (ret < 0)
 		return ret;
 
Signature · 1 line
2.47.3
#23 lore raw

Re: [PATCH v5 06/18] iio: magnetometer: ak8975: pass conversion timeouts as arguments

Quoted 1 line · On Tue, May 05, 2026 at 01:46:02PM +0200, Joshua Crofts via B4 Relay wrote:

Refactor wait_conversion_complete*_() helper function to accept poll

functions

Quoted 5 lines

and timeout values directly as parameters. This improves the readability of the code and does not rely on hardcoded macros.

Besides that, fix the home grown and obviously wrong in some cases the jiffy-based timeout.

...

drivers/iio/magnetometer/ak8975.c static int wait_conversion_complete_interrupt(struct ak8975_data *data) · line 718

Quoted 9 lines

static int ak8975_start_read_axis(struct ak8975_data *data,

drivers/iio/magnetometer/ak8975.c +3 -2
 				  const struct i2c_client *client)
 {
-	/* Set up the device for taking a sample. */
-	int ret = ak8975_set_mode(data, MODE_ONCE);
+	int ret;
 
+	/* Set up the device for taking a sample. */
+	ret = ak8975_set_mode(data, MODE_ONCE);

But this piece does not belong to the main scope of the patch.

Quoted 3 lines
drivers/iio/magnetometer/ak8975.c +0 -0
 	if (ret < 0) {
 		dev_err(&client->dev, "Error in setting operating mode\n");
 		return ret;
Signature · 2 lines
With Best Regards,
Andy Shevchenko
#24 lore raw

Re: [PATCH v5 06/18] iio: magnetometer: ak8975: pass conversion timeouts as arguments

Quoted 22 lines · Andy Shevchenko
Quoted 1 line · On Tue, May 05, 2026 at 01:46:02PM +0200, Joshua Crofts via B4 Relay wrote:

Refactor wait_conversion_complete*_() helper function to accept poll

functions

Quoted 5 lines

and timeout values directly as parameters. This improves the readability of the code and does not rely on hardcoded macros.

Besides that, fix the home grown and obviously wrong in some cases the jiffy-based timeout.

...

Quoted 7 lines

static int ak8975_start_read_axis(struct ak8975_data *data,
const struct i2c_client *client)
{

drivers/iio/magnetometer/ak8975.c +1 -2
-     /* Set up the device for taking a sample. */
-     int ret = ak8975_set_mode(data, MODE_ONCE);
+     int ret;
Quoted 4 lines

drivers/iio/magnetometer/ak8975.c static int wait_conversion_complete_interrupt(struct ak8975_data *data) · line 718

Quoted 2 lines
drivers/iio/magnetometer/ak8975.c +2 -0
+     /* Set up the device for taking a sample. */
+     ret = ak8975_set_mode(data, MODE_ONCE);

But this piece does not belong to the main scope of the patch.

Fair, probably a badly resolved merge conflict.

Signature · 3 lines
Kind regards

CJD
#25 lore raw

Re: [PATCH v5 06/18] iio: magnetometer: ak8975: pass conversion timeouts as arguments

Quoted 3 lines · Joshua Crofts
Quoted 1 line · Andy Shevchenko

On Tue, May 05, 2026 at 01:46:02PM +0200, Joshua Crofts via B4 Relay wrote:

...

Quoted 7 lines
Quoted 7 lines
Quoted 7 lines

static int ak8975_start_read_axis(struct ak8975_data *data,
const struct i2c_client *client)
{

drivers/iio/magnetometer/ak8975.c +1 -2
-     /* Set up the device for taking a sample. */
-     int ret = ak8975_set_mode(data, MODE_ONCE);
+     int ret;
Quoted 6 lines
Quoted 4 lines

drivers/iio/magnetometer/ak8975.c static int wait_conversion_complete_interrupt(struct ak8975_data *data) · line 718

Quoted 2 lines
drivers/iio/magnetometer/ak8975.c +2 -0
+     /* Set up the device for taking a sample. */
+     ret = ak8975_set_mode(data, MODE_ONCE);

But this piece does not belong to the main scope of the patch.

Fair, probably a badly resolved merge conflict.

Hmm... Usually what we use is interactive rebase. Very powerful tool!

	`git rebase --rebase-merges --interactive $SHA`

where $SHA is the same one that you used as --base for the series.

In conjunction with

	`git reset HEAD~1; git commit -s --interactive -c ORIG_HEAD`

technique, you can easily move parts between patches.

(There are even more advanced techniques, but I'm an old-school guy.)

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#26 lore raw

Re: [PATCH v5 06/18] iio: magnetometer: ak8975: pass conversion timeouts as arguments

Quoted 18 lines · Andy Shevchenko
Quoted 3 lines · Joshua Crofts
Quoted 1 line · Andy Shevchenko

On Tue, May 05, 2026 at 01:46:02PM +0200, Joshua Crofts via B4 Relay wrote:

...

Quoted 10 lines
Quoted 10 lines
Quoted 9 lines

static int ak8975_start_read_axis(struct ak8975_data *data,
const struct i2c_client *client)
{

drivers/iio/magnetometer/ak8975.c +3 -2
-     /* Set up the device for taking a sample. */
-     int ret = ak8975_set_mode(data, MODE_ONCE);
+     int ret;
 
+     /* Set up the device for taking a sample. */
+     ret = ak8975_set_mode(data, MODE_ONCE);
Quoted 7 lines
Quoted 3 lines
Quoted 1 line

But this piece does not belong to the main scope of the patch.

Fair, probably a badly resolved merge conflict.

Hmm... Usually what we use is interactive rebase. Very powerful tool!

        `git rebase --rebase-merges --interactive $SHA`

I see, well I use git rebase -i HEAD~no_of_commits_in_series and sometimes I move a different thing by accident. It does get a bit crowded with 18 patches (yours were fine but I also had to do some resolution after moving the order)

Quoted 9 lines

where $SHA is the same one that you used as --base for the series.

In conjunction with

        `git reset HEAD~1; git commit -s --interactive -c ORIG_HEAD`

technique, you can easily move parts between patches.

(There are even more advanced techniques, but I'm an old-school guy.)

Good to know, thanks!

Signature · 3 lines
Kind regards

CJD
#27 lore raw

Re: [PATCH v5 06/18] iio: magnetometer: ak8975: pass conversion timeouts as arguments

Quoted 6 lines · Joshua Crofts
Quoted 4 lines · Andy Shevchenko
Quoted 3 lines · Joshua Crofts
Quoted 1 line · Andy Shevchenko

On Tue, May 05, 2026 at 01:46:02PM +0200, Joshua Crofts via B4 Relay wrote:

...

Quoted 7 lines
Quoted 5 lines
Quoted 1 line

Fair, probably a badly resolved merge conflict.

Hmm... Usually what we use is interactive rebase. Very powerful tool!

        `git rebase --rebase-merges --interactive $SHA`

I see, well I use git rebase -i HEAD~no_of_commits_in_series

Good that you are already onboard!

Quoted 1 line

and sometimes I move a different thing by accident.

Yeah, this is usually unrelated to the tools, human factor, you know :-)

Quoted 8 lines

It does get a bit
crowded with 18 patches (yours were fine but I also had to do some resolution after moving the order)

Quoted 4 lines

where $SHA is the same one that you used as --base for the series.

In conjunction with

Quoted 7 lines
Quoted 5 lines
        `git reset HEAD~1; git commit -s --interactive -c ORIG_HEAD`

technique, you can easily move parts between patches.

(There are even more advanced techniques, but I'm an old-school guy.)

Good to know, thanks!

You're welcome!

Signature · 2 lines
With Best Regards,
Andy Shevchenko
maintainer
#28 lore raw

Re: [PATCH v5 06/18] iio: magnetometer: ak8975: pass conversion timeouts as arguments

Quoted 34 lines · On Wed, 6 May 2026 10:32:49 +0300 Andy Shevchenko…
Quoted 6 lines · Joshua Crofts
Quoted 4 lines · Andy Shevchenko
Quoted 3 lines · Joshua Crofts
Quoted 1 line · Andy Shevchenko

On Tue, May 05, 2026 at 01:46:02PM +0200, Joshua Crofts via B4 Relay wrote:

...

Quoted 7 lines
Quoted 5 lines
Quoted 1 line

Fair, probably a badly resolved merge conflict.

Hmm... Usually what we use is interactive rebase. Very powerful tool!

        `git rebase --rebase-merges --interactive $SHA`  

I see, well I use git rebase -i HEAD~no_of_commits_in_series

Good that you are already onboard!

Quoted 1 line

and sometimes I move a different thing by accident.

Yeah, this is usually unrelated to the tools, human factor, you know :-)

Quoted 10 lines

It does get a bit
crowded with 18 patches (yours were fine but I also had to do some resolution after moving the order)

Quoted 6 lines

where $SHA is the same one that you used as --base for the series.

In conjunction with

        `git reset HEAD~1; git commit -s --interactive -c ORIG_HEAD`
Quoted 8 lines
Quoted 5 lines
Quoted 3 lines

technique, you can easily move parts between patches.

(There are even more advanced techniques, but I'm an old-school guy.)

Good to know, thanks!

You're welcome!

Given the issue was minor I applied and dropped that change that should have been elsewhere.

Thanks,

Jonathan

#29 lore raw

PATCH v5 7/18 iio: magnetometer: ak8975: modernize polling loops with iopoll() macros

In reply to 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. Add <linux/time.h> for USEC_PER_MSEC macro instead of using magic numbers.

Assisted-by Gemini:3.1-Pro
---
 drivers/iio/magnetometer/ak8975.c | 43 +++++++++++++++++----------------------
 1 file changed, 19 insertions(+), 24 deletions(-)
drivers/iio/magnetometer/ak8975.c +19 -24
index 83878ea28b91d8316d2cc31e57fde6da957af689..129c98c97071b4cf8efa3abf2364a3abbde129ba 100644
@@ -24,6 +24,7 @@
 #include <linux/pm_runtime.h>
 #include <linux/property.h>
 #include <linux/regulator/consumer.h>
+#include <linux/time.h>
 #include <linux/types.h>
 #include <linux/wait.h>
 
@@ -649,16 +650,14 @@ static int wait_conversion_complete_gpio(struct ak8975_data *data,
 {
 	struct i2c_client *client = data->client;
 	int ret;
+	int val;
 
 	/* Wait for the conversion to complete. */
-	while (timeout_ms) {
-		msleep(poll_ms);
-		if (gpiod_get_value(data->eoc_gpiod))
-			break;
-		timeout_ms -= poll_ms;
-	}
-	if (!timeout_ms)
-		return -ETIMEDOUT;
+	ret = readx_poll_timeout(gpiod_get_value, data->eoc_gpiod, val, val != 0,
+				 poll_ms * USEC_PER_MSEC,
+				 timeout_ms * USEC_PER_MSEC);
+	if (ret)
+		return ret;
 
 	ret = i2c_smbus_read_byte_data(client, data->def->ctrl_regs[ST1]);
 	if (ret < 0)
@@ -672,27 +671,23 @@ static int wait_conversion_complete_polled(struct ak8975_data *data,
 					   unsigned int timeout_ms)
 {
 	struct i2c_client *client = data->client;
-	u8 read_status;
 	int ret;
+	int val;
 
 	/* Wait for the conversion to complete. */
-	while (timeout_ms) {
-		msleep(poll_ms);
-		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 -= poll_ms;
+	ret = read_poll_timeout(i2c_smbus_read_byte_data, val, val != 0,
+				poll_ms * USEC_PER_MSEC,
+				timeout_ms * USEC_PER_MSEC,
+				true,
+				client, data->def->ctrl_regs[ST1]);
+	if (ret)
+		return ret;
+	if (val < 0) {
+		dev_err(&client->dev, "Error in reading ST1\n");
+		return val;
 	}
-	if (!timeout_ms)
-		return -ETIMEDOUT;
 
-	return read_status;
+	return val;
 }
 
 /* Returns 0 if the end of conversion interrupt occurred or -ETIMEDOUT otherwise */
Signature · 1 line
2.47.3
#30 lore raw

Re: [PATCH v5 07/18] iio: magnetometer: ak8975: modernize polling loops with iopoll() macros

Quoted 8 lines · On Tue, May 05, 2026 at 01:46:03PM +0200, Joshua Crofts via B4 Relay 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. Add <linux/time.h> for USEC_PER_MSEC macro instead of using magic numbers.

...

Quoted 6 lines
drivers/iio/magnetometer/ak8975.c +6 -0
+	ret = read_poll_timeout(i2c_smbus_read_byte_data, val, val != 0,
+				poll_ms * USEC_PER_MSEC,
+				timeout_ms * USEC_PER_MSEC,
+				true,
+				client, data->def->ctrl_regs[ST1]);
+	if (ret)

drivers/iio/magnetometer/ak8975.c static int wait_conversion_complete_polled(struct ak8975_data *data, · line 690

Quoted 10 lines
drivers/iio/magnetometer/ak8975.c +5 -3
+		return ret;
+	if (val < 0) {
+		dev_err(&client->dev, "Error in reading ST1\n");
+		return val;
 	}
-	if (!timeout_ms)
-		return -ETIMEDOUT;
 
-	return read_status;
+	return val;

Besides the unneeded return val duplication, I think this is the wrong location for this check and it changes behaviour (really subtle change!).

Before if the last iteration gives an error from the device, we return -ETIMEDOUT instead of the whatever the i2c_smbus_read_byte_data() returns.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#31 lore raw

Re: [PATCH v5 07/18] iio: magnetometer: ak8975: modernize polling loops with iopoll() macros

Quoted 28 lines · Andy Shevchenko
Quoted 8 lines · On Tue, May 05, 2026 at 01:46:03PM +0200, Joshua Crofts via B4 Relay 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. Add <linux/time.h> for USEC_PER_MSEC macro instead of using magic numbers.

...

Quoted 14 lines
drivers/iio/magnetometer/ak8975.c +10 -2
+     ret = read_poll_timeout(i2c_smbus_read_byte_data, val, val != 0,
+                             poll_ms * USEC_PER_MSEC,
+                             timeout_ms * USEC_PER_MSEC,
+                             true,
+                             client, data->def->ctrl_regs[ST1]);
+     if (ret)
+             return ret;
+     if (val < 0) {
+             dev_err(&client->dev, "Error in reading ST1\n");
+             return val;
      }
-     if (!timeout_ms)
-             return -ETIMEDOUT;
Quoted 8 lines

drivers/iio/magnetometer/ak8975.c static int wait_conversion_complete_polled(struct ak8975_data *data, · line 690

Quoted 2 lines
drivers/iio/magnetometer/ak8975.c +1 -1
-     return read_status;
+     return val;

Besides the unneeded return val duplication, I think this is the wrong location for this check and it changes behaviour (really subtle change!).

Before if the last iteration gives an error from the device, we return -ETIMEDOUT instead of the whatever the i2c_smbus_read_byte_data() returns.

Yeah, that is pretty subtle, but I agree.

Signature · 3 lines
Kind regards

CJD
#32 lore raw

Re: [PATCH v5 07/18] iio: magnetometer: ak8975: modernize polling loops with iopoll() macros

Quoted 31 lines · Joshua Crofts
Quoted 28 lines · Andy Shevchenko
Quoted 8 lines · On Tue, May 05, 2026 at 01:46:03PM +0200, Joshua Crofts via B4 Relay 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. Add <linux/time.h> for USEC_PER_MSEC macro instead of using magic numbers.

...

Quoted 14 lines
drivers/iio/magnetometer/ak8975.c +10 -2
+     ret = read_poll_timeout(i2c_smbus_read_byte_data, val, val != 0,
+                             poll_ms * USEC_PER_MSEC,
+                             timeout_ms * USEC_PER_MSEC,
+                             true,
+                             client, data->def->ctrl_regs[ST1]);
+     if (ret)
+             return ret;
+     if (val < 0) {
+             dev_err(&client->dev, "Error in reading ST1\n");
+             return val;
      }
-     if (!timeout_ms)
-             return -ETIMEDOUT;
Quoted 5 lines
Quoted 5 lines

drivers/iio/magnetometer/ak8975.c static int wait_conversion_complete_polled(struct ak8975_data *data, · line 690

Quoted 2 lines
drivers/iio/magnetometer/ak8975.c +1 -1
-     return read_status;
+     return val;

Besides the unneeded return val duplication, I think this is the wrong location for this check and it changes behaviour (really subtle change!).

Hmm, rechecking this, I agree with the unnecessary return, but I don't think it changes the behaviour of the function.

Quoted 2 lines
Quoted 2 lines

Before if the last iteration gives an error from the device, we return -ETIMEDOUT instead of the whatever the i2c_smbus_read_byte_data() returns.

Looking at the original function, it always prioritized returning the i2c_smbus_read_byte_data() error code before returning -ETIMEDOUT (even on the final iteration). In the new version, if the i2c read is bad, the read_poll_timeout()
code will still be zero, therefore allowing the code to jump to the i2c value check
and then return if bad (this is still the same behaviour IMO).

Maybe I'm looking at it from the wrong angle though. AFAIC Sashiko nitpicked everything except this (but yes, it's still an AI so we have to tread lightly).

Same goes for the other patch where you raised this issue.

Signature · 3 lines
Kind regards

CJD
#33 lore raw

Re: [PATCH v5 07/18] iio: magnetometer: ak8975: modernize polling loops with iopoll() macros

Quoted 4 lines · Joshua Crofts
Quoted 3 lines · Joshua Crofts
Quoted 1 line · Andy Shevchenko

On Tue, May 05, 2026 at 01:46:03PM +0200, Joshua Crofts via B4 Relay wrote:

...

Quoted 26 lines
Quoted 14 lines
Quoted 14 lines
Quoted 14 lines
drivers/iio/magnetometer/ak8975.c +10 -2
+     ret = read_poll_timeout(i2c_smbus_read_byte_data, val, val != 0,
+                             poll_ms * USEC_PER_MSEC,
+                             timeout_ms * USEC_PER_MSEC,
+                             true,
+                             client, data->def->ctrl_regs[ST1]);
+     if (ret)
+             return ret;
+     if (val < 0) {
+             dev_err(&client->dev, "Error in reading ST1\n");
+             return val;
      }
-     if (!timeout_ms)
-             return -ETIMEDOUT;
Quoted 5 lines
Quoted 5 lines

drivers/iio/magnetometer/ak8975.c static int wait_conversion_complete_polled(struct ak8975_data *data, · line 690

Quoted 2 lines
drivers/iio/magnetometer/ak8975.c +1 -1
-     return read_status;
+     return val;

Besides the unneeded return val duplication, I think this is the wrong location for this check and it changes behaviour (really subtle change!).

Hmm, rechecking this, I agree with the unnecessary return, but I don't think it changes the behaviour of the function.

Quoted 2 lines
Quoted 2 lines

Before if the last iteration gives an error from the device, we return -ETIMEDOUT instead of the whatever the i2c_smbus_read_byte_data() returns.

Quoted 7 lines

Looking at the original function, it always prioritized returning the i2c_smbus_read_byte_data() error code before returning -ETIMEDOUT (even on the final iteration). In the new version, if the i2c read is bad, the read_poll_timeout()
code will still be zero, therefore allowing the code to jump to the i2c value check
and then return if bad (this is still the same behaviour IMO).

It looks like I stand corrected! Indeed, we have two conditions to follow, one is provided in a macro parameter, and the other is for timeout. Here the Q is which one logically should be checked first? More thinking on it tends to your direction as it follows usual pattern as we check the return values (errors) from the callee first *then* validate the results.

Quoted 4 lines

Maybe I'm looking at it from the wrong angle though. AFAIC Sashiko nitpicked everything except this (but yes, it's still an AI so we have to tread lightly).

Same goes for the other patch where you raised this issue.

Yep.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
maintainer
#34 lore raw

Re: [PATCH v5 07/18] iio: magnetometer: ak8975: modernize polling loops with iopoll() macros

Quoted 48 lines · On Wed, 6 May 2026 16:08:21 +0300 Andy Shevchenko…
Quoted 4 lines · Joshua Crofts
Quoted 3 lines · Joshua Crofts
Quoted 1 line · Andy Shevchenko

On Tue, May 05, 2026 at 01:46:03PM +0200, Joshua Crofts via B4 Relay wrote:

...

Quoted 26 lines
Quoted 14 lines
Quoted 14 lines
Quoted 14 lines
drivers/iio/magnetometer/ak8975.c +10 -2
+     ret = read_poll_timeout(i2c_smbus_read_byte_data, val, val != 0,
+                             poll_ms * USEC_PER_MSEC,
+                             timeout_ms * USEC_PER_MSEC,
+                             true,
+                             client, data->def->ctrl_regs[ST1]);
+     if (ret)
+             return ret;
+     if (val < 0) {
+             dev_err(&client->dev, "Error in reading ST1\n");
+             return val;
      }
-     if (!timeout_ms)
-             return -ETIMEDOUT;
Quoted 5 lines
Quoted 5 lines

drivers/iio/magnetometer/ak8975.c static int wait_conversion_complete_polled(struct ak8975_data *data, · line 690

Quoted 2 lines
drivers/iio/magnetometer/ak8975.c +1 -1
-     return read_status;
+     return val;  

Besides the unneeded return val duplication, I think this is the wrong location for this check and it changes behaviour (really subtle change!).

Hmm, rechecking this, I agree with the unnecessary return, but I don't think it changes the behaviour of the function.

Quoted 2 lines
Quoted 2 lines

Before if the last iteration gives an error from the device, we return -ETIMEDOUT instead of the whatever the i2c_smbus_read_byte_data() returns.

Quoted 7 lines

Looking at the original function, it always prioritized returning the i2c_smbus_read_byte_data() error code before returning -ETIMEDOUT (even on the final iteration). In the new version, if the i2c read is bad, the read_poll_timeout()
code will still be zero, therefore allowing the code to jump to the i2c value check
and then return if bad (this is still the same behaviour IMO).

It looks like I stand corrected! Indeed, we have two conditions to follow, one is provided in a macro parameter, and the other is for timeout. Here the Q is which one logically should be checked first? More thinking on it tends to your direction as it follows usual pattern as we check the return values (errors) from the callee first *then* validate the results.

Quoted 7 lines
Quoted 4 lines

Maybe I'm looking at it from the wrong angle though. AFAIC Sashiko nitpicked everything except this (but yes, it's still an AI so we have to tread lightly).

Same goes for the other patch where you raised this issue.

Yep.

I'll leave this one for now as the return val duplication can be tidied up. I think the rest of the patch looks good tome.

Thanks,

Jonathan

#35 lore raw

PATCH v5 8/18 iio: magnetometer: ak8975: check if gpiod read was successful

In reply to Joshua Crofts

Add a check that ensures that valid data has been read from GPIOD. If not, log an error and return the negative read value.

---
 drivers/iio/magnetometer/ak8975.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/iio/magnetometer/ak8975.c +4 -0
index 129c98c97071b4cf8efa3abf2364a3abbde129ba..b3caf5f3ebf29bf701bb214aa480fa7afb2651c6 100644
@@ -658,6 +658,10 @@ static int wait_conversion_complete_gpio(struct ak8975_data *data,
 				 timeout_ms * USEC_PER_MSEC);
 	if (ret)
 		return ret;
+	if (val < 0) {
+		dev_err(&client->dev, "Error in reading GPIOD\n");
+		return val;
+	}
 
 	ret = i2c_smbus_read_byte_data(client, data->def->ctrl_regs[ST1]);
 	if (ret < 0)
Signature · 1 line
2.47.3
Reviewed-by on v5 8/18
#36 lore raw

Re: [PATCH v5 08/18] iio: magnetometer: ak8975: check if gpiod read was successful

Quoted 16 lines · On Tue May 5, 2026 at 6:46 AM CDT, Joshua Crofts via B4 Relay wrote:

From: Joshua Crofts <[email protected]>

Add a check that ensures that valid data has been read from GPIOD. If not, log an error and return the negative read value.

---
 drivers/iio/magnetometer/ak8975.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/iio/magnetometer/ak8975.c +0 -0
index 129c98c97071b4cf8efa3abf2364a3abbde129ba..b3caf5f3ebf29bf701bb214aa480fa7afb2651c6 100644

drivers/iio/magnetometer/ak8975.c static int wait_conversion_complete_gpio(struct ak8975_data *data, · line 667

Quoted 10 lines
drivers/iio/magnetometer/ak8975.c +4 -0
@@ -658,6 +658,10 @@ static int wait_conversion_complete_gpio(struct ak8975_data *data,
 				 timeout_ms * USEC_PER_MSEC);
 	if (ret)
 		return ret;
+	if (val < 0) {
+		dev_err(&client->dev, "Error in reading GPIOD\n");
+		return val;
+	}
 
 	ret = i2c_smbus_read_byte_data(client, data->def->ctrl_regs[ST1]);
 	if (ret < 0)

Small nit but I wonder if we can change:

	if (val < 0) {
		dev_err(&client->dev, "Error in reading GPIOD\n");
		return val;
	}

to something like:

	if (val < 0)
		return dev_err_probe(&client->dev, val, "Error in reading GPIOD\n");

Would be good for code density.

My rb will apply either way.

best regards,
max

#37 lore raw

Re: [PATCH v5 08/18] iio: magnetometer: ak8975: check if gpiod read was successful

Quoted 41 lines · Maxwell Doose
Quoted 16 lines · On Tue May 5, 2026 at 6:46 AM CDT, Joshua Crofts via B4 Relay wrote:

From: Joshua Crofts <[email protected]>

Add a check that ensures that valid data has been read from GPIOD. If not, log an error and return the negative read value.

---
 drivers/iio/magnetometer/ak8975.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/iio/magnetometer/ak8975.c +0 -0
index 129c98c97071b4cf8efa3abf2364a3abbde129ba..b3caf5f3ebf29bf701bb214aa480fa7afb2651c6 100644

drivers/iio/magnetometer/ak8975.c static int wait_conversion_complete_gpio(struct ak8975_data *data, · line 667

Quoted 10 lines
drivers/iio/magnetometer/ak8975.c +4 -0
@@ -658,6 +658,10 @@ static int wait_conversion_complete_gpio(struct ak8975_data *data,
                               timeout_ms * USEC_PER_MSEC);
      if (ret)
              return ret;
+     if (val < 0) {
+             dev_err(&client->dev, "Error in reading GPIOD\n");
+             return val;
+     }
 
      ret = i2c_smbus_read_byte_data(client, data->def->ctrl_regs[ST1]);
      if (ret < 0)

Small nit but I wonder if we can change:

        if (val < 0) {
                dev_err(&client->dev, "Error in reading GPIOD\n");
                return val;
        }

to something like:

        if (val < 0)
                return dev_err_probe(&client->dev, val, "Error in reading GPIOD\n");
Quoted 8 lines

Would be good for code density.

My rb will apply either way.

best regards,
max

Actually, timeout. Sashiko had a good potential catch here:

https://sashiko.dev/#/patchset/20260505-magnetometer-fixes-v5-0-831b9b5550fc%40gmail.com

"Does this new error path leak the pm runtime reference?
In ak8975_read_axis(), a pm runtime reference is acquired via pm_runtime_get_sync(&data->client->dev). If this new error path returns val back up the call chain through ak8975_start_read_axis() to ak8975_read_axis(), the code jumps to the exit label.
The exit block unlocks the mutex and returns the error code, but it skips calling pm_runtime_put_autosuspend()."

I'm going to withdraw my reviewed-by until I can confirm that this isn't a problem, sorry about that.

best regards,
max

#38 lore raw

Re: [PATCH v5 08/18] iio: magnetometer: ak8975: check if gpiod read was successful

Quoted 2 lines · Maxwell Doose
Quoted 1 line · Maxwell Doose

On Tue May 5, 2026 at 6:46 AM CDT, Joshua Crofts via B4 Relay wrote:

...

Quoted 11 lines

Actually, timeout. Sashiko had a good potential catch here:

https://sashiko.dev/#/patchset/20260505-magnetometer-fixes-v5-0-831b9b5550fc%40gmail.com

"Does this new error path leak the pm runtime reference?
In ak8975_read_axis(), a pm runtime reference is acquired via pm_runtime_get_sync(&data->client->dev). If this new error path returns val back up the call chain through ak8975_start_read_axis() to ak8975_read_axis(), the code jumps to the exit label.
The exit block unlocks the mutex and returns the error code, but it skips calling pm_runtime_put_autosuspend()."

While it seems like a valid concern, it doesn't relate to this change.

Quoted 2 lines

I'm going to withdraw my reviewed-by until I can confirm that this isn't a problem, sorry about that.

Why?

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#39 lore raw

Re: [PATCH v5 08/18] iio: magnetometer: ak8975: check if gpiod read was successful

Quoted 19 lines · Andy Shevchenko
Quoted 2 lines · Maxwell Doose
Quoted 1 line · Maxwell Doose

On Tue May 5, 2026 at 6:46 AM CDT, Joshua Crofts via B4 Relay wrote:

...

Quoted 11 lines

Actually, timeout. Sashiko had a good potential catch here:

https://sashiko.dev/#/patchset/20260505-magnetometer-fixes-v5-0-831b9b5550fc%40gmail.com

"Does this new error path leak the pm runtime reference?
In ak8975_read_axis(), a pm runtime reference is acquired via pm_runtime_get_sync(&data->client->dev). If this new error path returns val back up the call chain through ak8975_start_read_axis() to ak8975_read_axis(), the code jumps to the exit label.
The exit block unlocks the mutex and returns the error code, but it skips calling pm_runtime_put_autosuspend()."

Quoted 7 lines

While it seems like a valid concern, it doesn't relate to this change.

Quoted 2 lines

I'm going to withdraw my reviewed-by until I can confirm that this isn't a problem, sorry about that.

Why?

I just wanted to make sure this wouldn't cause some crazy regression, I think it's valid to be somewhat concerned. If you want you can reapply my tag.

best regards,
max

Quoted 5 lines

--
With Best Regards,
Andy Shevchenko

#40 lore raw

Re: [PATCH v5 08/18] iio: magnetometer: ak8975: check if gpiod read was successful

Quoted 5 lines · Maxwell Doose
Quoted 3 lines · Andy Shevchenko
Quoted 2 lines · Maxwell Doose
Quoted 1 line · Maxwell Doose

On Tue May 5, 2026 at 6:46 AM CDT, Joshua Crofts via B4 Relay wrote:

...

Quoted 14 lines
Quoted 14 lines
Quoted 11 lines

Actually, timeout. Sashiko had a good potential catch here:

https://sashiko.dev/#/patchset/20260505-magnetometer-fixes-v5-0-831b9b5550fc%40gmail.com

"Does this new error path leak the pm runtime reference?
In ak8975_read_axis(), a pm runtime reference is acquired via pm_runtime_get_sync(&data->client->dev). If this new error path returns val back up the call chain through ak8975_start_read_axis() to ak8975_read_axis(), the code jumps to the exit label.
The exit block unlocks the mutex and returns the error code, but it skips calling pm_runtime_put_autosuspend()."

While it seems like a valid concern, it doesn't relate to this change.

Quoted 7 lines
Quoted 4 lines
Quoted 2 lines

I'm going to withdraw my reviewed-by until I can confirm that this isn't a problem, sorry about that.

Why?

I just wanted to make sure this wouldn't cause some crazy regression, I think it's valid to be somewhat concerned.

Yes, but this was the concern even before this patch, no?

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#41 lore raw

Re: [PATCH v5 08/18] iio: magnetometer: ak8975: check if gpiod read was successful

Quoted 7 lines · Andy Shevchenko
Quoted 5 lines · Maxwell Doose
Quoted 3 lines · Andy Shevchenko
Quoted 2 lines · Maxwell Doose
Quoted 1 line · Maxwell Doose

On Tue May 5, 2026 at 6:46 AM CDT, Joshua Crofts via B4 Relay wrote:

[snip]

Quoted 10 lines
Quoted 7 lines
Quoted 4 lines
Quoted 2 lines

I'm going to withdraw my reviewed-by until I can confirm that this isn't a problem, sorry about that.

Why?

I just wanted to make sure this wouldn't cause some crazy regression, I think it's valid to be somewhat concerned.

Yes, but this was the concern even before this patch, no?

Going to give that a solid maybe. I must've been tired when I looked over this again. Very sorry about that.

best regards,
max

Quoted 7 lines

--
With Best Regards,
Andy Shevchenko

#42 lore raw

Re: [PATCH v5 08/18] iio: magnetometer: ak8975: check if gpiod read was successful

In reply to Andy Shevchenko
Quoted 27 lines · Andy Shevchenko
Quoted 5 lines · Maxwell Doose
Quoted 3 lines · Andy Shevchenko
Quoted 2 lines · Maxwell Doose
Quoted 1 line · Maxwell Doose

On Tue May 5, 2026 at 6:46 AM CDT, Joshua Crofts via B4 Relay wrote:

...

Quoted 17 lines
Quoted 17 lines
Quoted 11 lines

Actually, timeout. Sashiko had a good potential catch here:

https://sashiko.dev/#/patchset/20260505-magnetometer-fixes-v5-0-831b9b5550fc%40gmail.com

"Does this new error path leak the pm runtime reference?
In ak8975_read_axis(), a pm runtime reference is acquired via pm_runtime_get_sync(&data->client->dev). If this new error path returns val back up the call chain through ak8975_start_read_axis() to ak8975_read_axis(), the code jumps to the exit label.
The exit block unlocks the mutex and returns the error code, but it skips calling pm_runtime_put_autosuspend()."

While it seems like a valid concern, it doesn't relate to this change.

Quoted 2 lines

I'm going to withdraw my reviewed-by until I can confirm that this isn't a problem, sorry about that.

Quoted 6 lines
Quoted 4 lines
Quoted 1 line

Why?

I just wanted to make sure this wouldn't cause some crazy regression, I think it's valid to be somewhat concerned.

Yes, but this was the concern even before this patch, no?

IMO yes, even the first error path would leak the pm_runtime reference.

Signature · 3 lines
Kind regards

CJD
#43 lore raw

Re: [PATCH v5 08/18] iio: magnetometer: ak8975: check if gpiod read was successful

In reply to Maxwell Doose
Quoted 36 lines · Maxwell Doose
Quoted 16 lines · On Tue May 5, 2026 at 6:46 AM CDT, Joshua Crofts via B4 Relay wrote:

From: Joshua Crofts <[email protected]>

Add a check that ensures that valid data has been read from GPIOD. If not, log an error and return the negative read value.

---
 drivers/iio/magnetometer/ak8975.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/iio/magnetometer/ak8975.c +0 -0
index 129c98c97071b4cf8efa3abf2364a3abbde129ba..b3caf5f3ebf29bf701bb214aa480fa7afb2651c6 100644

drivers/iio/magnetometer/ak8975.c static int wait_conversion_complete_gpio(struct ak8975_data *data, · line 667

Quoted 10 lines
drivers/iio/magnetometer/ak8975.c +4 -0
@@ -658,6 +658,10 @@ static int wait_conversion_complete_gpio(struct ak8975_data *data,
                               timeout_ms * USEC_PER_MSEC);
      if (ret)
              return ret;
+     if (val < 0) {
+             dev_err(&client->dev, "Error in reading GPIOD\n");
+             return val;
+     }
 
      ret = i2c_smbus_read_byte_data(client, data->def->ctrl_regs[ST1]);
      if (ret < 0)

Small nit but I wonder if we can change:

        if (val < 0) {
                dev_err(&client->dev, "Error in reading GPIOD\n");
                return val;
        }
Quoted 4 lines

to something like:

        if (val < 0)
                return dev_err_probe(&client->dev, val, "Error in reading GPIOD\n");

Sure, but it would go as a separate patch. Given this series is 18 parts long, I'd rather get the existing patches merged and work on this in another series. This driver still has a few shortcomings that should be ironed out.

Signature · 3 lines
Kind regards

CJD
maintainer
#44 lore raw

Re: [PATCH v5 08/18] iio: magnetometer: ak8975: check if gpiod read was successful

Quoted 39 lines · On Wed, 6 May 2026 09:05:39 +0200 Joshua Crofts <[email protected]>…
Quoted 38 lines · Maxwell Doose
Quoted 16 lines · On Tue May 5, 2026 at 6:46 AM CDT, Joshua Crofts via B4 Relay wrote:

From: Joshua Crofts <[email protected]>

Add a check that ensures that valid data has been read from GPIOD. If not, log an error and return the negative read value.

---
 drivers/iio/magnetometer/ak8975.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/iio/magnetometer/ak8975.c +0 -0
index 129c98c97071b4cf8efa3abf2364a3abbde129ba..b3caf5f3ebf29bf701bb214aa480fa7afb2651c6 100644

drivers/iio/magnetometer/ak8975.c static int wait_conversion_complete_gpio(struct ak8975_data *data, · line 667

Quoted 10 lines
drivers/iio/magnetometer/ak8975.c +4 -0
@@ -658,6 +658,10 @@ static int wait_conversion_complete_gpio(struct ak8975_data *data,
                               timeout_ms * USEC_PER_MSEC);
      if (ret)
              return ret;
+     if (val < 0) {
+             dev_err(&client->dev, "Error in reading GPIOD\n");
+             return val;
+     }
 
      ret = i2c_smbus_read_byte_data(client, data->def->ctrl_regs[ST1]);
      if (ret < 0)  

Small nit but I wonder if we can change:

        if (val < 0) {
                dev_err(&client->dev, "Error in reading GPIOD\n");
                return val;
        }

to something like:

Quoted 8 lines
Quoted 2 lines
        if (val < 0)
                return dev_err_probe(&client->dev, val, "Error in reading GPIOD\n");  

Sure, but it would go as a separate patch. Given this series is 18 parts long, I'd rather get the existing patches merged and work on this in another series. This driver still has a few shortcomings that should be ironed out.

This doesn't look to only be called from probe. dev_err_probe() is only meant for stuff that is. Arguably you could use it elsewhere but then it should have a different name.

Thanks,

Jonathan

#45 lore raw

Re: [PATCH v5 08/18] iio: magnetometer: ak8975: check if gpiod read was successful

In reply to Maxwell Doose
Quoted 1 line · Maxwell Doose

On Tue May 5, 2026 at 6:46 AM CDT, Joshua Crofts via B4 Relay wrote:

...

Quoted 7 lines

Small nit but I wonder if we can change:

	if (val < 0) {
		dev_err(&client->dev, "Error in reading GPIOD\n");
		return val;
	}
Quoted 6 lines

to something like:

	if (val < 0)
		return dev_err_probe(&client->dev, val, "Error in reading GPIOD\n");

Would be good for code density.

Not in this patch. IIRC we have a dedicated change for dev_err_probe() conversion. Also note, dev_err_probe() is only applicable to the functions that are part of the probe stage and probe stage only. If the function shared between probe and run-time, the dev_err_probe() has not to be used.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#46 lore raw

Re: [PATCH v5 08/18] iio: magnetometer: ak8975: check if gpiod read was successful

Quoted 18 lines · Andy Shevchenko
Quoted 1 line · Maxwell Doose

On Tue May 5, 2026 at 6:46 AM CDT, Joshua Crofts via B4 Relay wrote:

...

Quoted 12 lines

Small nit but I wonder if we can change:

      if (val < 0) {
              dev_err(&client->dev, "Error in reading GPIOD\n");
              return val;
      }

to something like:

      if (val < 0)
              return dev_err_probe(&client->dev, val, "Error in reading GPIOD\n");
Quoted 7 lines
Quoted 1 line

Would be good for code density.

Not in this patch. IIRC we have a dedicated change for dev_err_probe() conversion. Also note, dev_err_probe() is only applicable to the functions that are part of the probe stage and probe stage only. If the function shared between probe and run-time, the dev_err_probe() has not to be used.

Understood; I'll take that into account for the future, thanks for that tip.

best regards,
max

Quoted 5 lines

--
With Best Regards,
Andy Shevchenko

#47 lore raw

Re: [PATCH v5 08/18] iio: magnetometer: ak8975: check if gpiod read was successful

In reply to Joshua Crofts
Quoted 2 lines · On Tue, May 05, 2026 at 01:46:04PM +0200, Joshua Crofts via B4 Relay wrote:

Add a check that ensures that valid data has been read from GPIOD. If not, log an error and return the negative read value.

...

See the comment on the previous patch, applies to here as well (to some extent).

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#48 lore raw

Re: [PATCH v5 08/18] iio: magnetometer: ak8975: check if gpiod read was successful

Quoted 10 lines · Andy Shevchenko
Quoted 2 lines · On Tue, May 05, 2026 at 01:46:04PM +0200, Joshua Crofts via B4 Relay wrote:

Add a check that ensures that valid data has been read from GPIOD. If not, log an error and return the negative read value.

...

See the comment on the previous patch, applies to here as well (to some extent).

Will do.

Signature · 3 lines
Kind regards

CJD
#49 lore raw

PATCH v5 9/18 iio: magnetometer: ak8975: avoid using temporary variable

In reply to Joshua Crofts

From: Andy Shevchenko <[email protected]>

Avoid using temporary variable in ak8975_read_axis(). With that being done, the clamp_t() call becomes idiomatic in the driver and can be factored out to a helper later on (and if needed).

---
 drivers/iio/magnetometer/ak8975.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/iio/magnetometer/ak8975.c +2 -3
index b3caf5f3ebf29bf701bb214aa480fa7afb2651c6..c36d068081066ef6b42cc2f9c3c2e8978ec51623 100644
@@ -741,7 +741,6 @@ static int ak8975_read_axis(struct iio_dev *indio_dev, int index, int *val)
 	const struct i2c_client *client = data->client;
 	const struct ak_def *def = data->def;
 	__le16 rval;
-	u16 buff;
 	int ret;
 
 	pm_runtime_get_sync(&data->client->dev);
@@ -778,8 +777,8 @@ static int ak8975_read_axis(struct iio_dev *indio_dev, int index, int *val)
 	pm_runtime_put_autosuspend(&data->client->dev);
 
 	/* Swap bytes and convert to valid range. */
-	buff = le16_to_cpu(rval);
-	*val = clamp_t(s16, buff, -def->range, def->range);
+	*val = clamp_t(s16, le16_to_cpu(rval), -def->range, def->range);
+
 	return IIO_VAL_INT;
 
 exit:
Signature · 1 line
2.47.3
maintainer
#50 lore raw

Re: [PATCH v5 09/18] iio: magnetometer: ak8975: avoid using temporary variable

Quoted 8 lines · On Tue, 05 May 2026 13:46:05 +0200 Joshua Crofts via B4 Relay…

From: Andy Shevchenko <[email protected]>

Avoid using temporary variable in ak8975_read_axis(). With that being done, the clamp_t() call becomes idiomatic in the driver and can be factored out to a helper later on (and if needed).

Independent of the last two patches (which I've left for now) so applied this one to the testing branch of iio.git.

Jonathan

Quoted 28 lines
---
 drivers/iio/magnetometer/ak8975.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/iio/magnetometer/ak8975.c +2 -3
index b3caf5f3ebf29bf701bb214aa480fa7afb2651c6..c36d068081066ef6b42cc2f9c3c2e8978ec51623 100644
@@ -741,7 +741,6 @@ static int ak8975_read_axis(struct iio_dev *indio_dev, int index, int *val)
 	const struct i2c_client *client = data->client;
 	const struct ak_def *def = data->def;
 	__le16 rval;
-	u16 buff;
 	int ret;
 
 	pm_runtime_get_sync(&data->client->dev);
@@ -778,8 +777,8 @@ static int ak8975_read_axis(struct iio_dev *indio_dev, int index, int *val)
 	pm_runtime_put_autosuspend(&data->client->dev);
 
 	/* Swap bytes and convert to valid range. */
-	buff = le16_to_cpu(rval);
-	*val = clamp_t(s16, buff, -def->range, def->range);
+	*val = clamp_t(s16, le16_to_cpu(rval), -def->range, def->range);
+
 	return IIO_VAL_INT;
 
 exit:
#51 lore raw

PATCH v5 10/18 iio: magnetometer: ak8975: drop duplicate NULL check

In reply to Joshua Crofts

From: Andy Shevchenko <[email protected]>

The gpiod_set_consumer_name() is NULL-aware, no need to perform the same check in the caller.

---
 drivers/iio/magnetometer/ak8975.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/iio/magnetometer/ak8975.c +1 -2
index c36d068081066ef6b42cc2f9c3c2e8978ec51623..7a971de2eebf127fc8ab68300fb82f38d9476c0c 100644
@@ -918,8 +918,7 @@ static int ak8975_probe(struct i2c_client *client)
 	eoc_gpiod = devm_gpiod_get_optional(&client->dev, NULL, GPIOD_IN);
 	if (IS_ERR(eoc_gpiod))
 		return PTR_ERR(eoc_gpiod);
-	if (eoc_gpiod)
-		gpiod_set_consumer_name(eoc_gpiod, "ak_8975");
+	gpiod_set_consumer_name(eoc_gpiod, "ak_8975");
 
 	/*
 	 * According to AK09911 datasheet, if reset GPIO is provided then
Signature · 1 line
2.47.3
maintainer Applied
#52 lore raw

Re: [PATCH v5 10/18] iio: magnetometer: ak8975: drop duplicate NULL check

Quoted 7 lines · On Tue, 05 May 2026 13:46:06 +0200 Joshua Crofts via B4 Relay…

From: Andy Shevchenko <[email protected]>

The gpiod_set_consumer_name() is NULL-aware, no need to perform the same check in the caller.

Applied.

Quoted 19 lines
---
 drivers/iio/magnetometer/ak8975.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/iio/magnetometer/ak8975.c +1 -2
index c36d068081066ef6b42cc2f9c3c2e8978ec51623..7a971de2eebf127fc8ab68300fb82f38d9476c0c 100644
@@ -918,8 +918,7 @@ static int ak8975_probe(struct i2c_client *client)
 	eoc_gpiod = devm_gpiod_get_optional(&client->dev, NULL, GPIOD_IN);
 	if (IS_ERR(eoc_gpiod))
 		return PTR_ERR(eoc_gpiod);
-	if (eoc_gpiod)
-		gpiod_set_consumer_name(eoc_gpiod, "ak_8975");
+	gpiod_set_consumer_name(eoc_gpiod, "ak_8975");
 
 	/*
 	 * According to AK09911 datasheet, if reset GPIO is provided then
#53 lore raw

PATCH v5 11/18 iio: magnetometer: ak8975: remove duplicate error message

In reply to Joshua Crofts

From: Andy Shevchenko <[email protected]>

The devm_request_irq() already prints an error message. Remove the duplicate.

---
 drivers/iio/magnetometer/ak8975.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)
drivers/iio/magnetometer/ak8975.c +1 -4
index 7a971de2eebf127fc8ab68300fb82f38d9476c0c..f58cbafb491a5e039b22fc928a6c58dd6ec0f297 100644
@@ -584,17 +584,14 @@ static int ak8975_setup_irq(struct ak8975_data *data)
 	rc = devm_request_irq(&client->dev, irq, ak8975_irq_handler,
 			      IRQF_TRIGGER_RISING,
 			      dev_name(&client->dev), data);
-	if (rc < 0) {
-		dev_err(&client->dev, "irq %d request failed: %d\n", irq, rc);
+	if (rc < 0)
 		return rc;
-	}
 
 	data->eoc_irq = irq;
 
 	return rc;
 }
 
-
 /*
  * Perform some start-of-day setup, including reading the asa calibration
  * values and caching them.
Signature · 1 line
2.47.3
#55 lore raw

PATCH v5 12/18 iio: magnetometer: ak8975: reduce usage of magic lengths of the buffer

In reply to Joshua Crofts

From: Andy Shevchenko <[email protected]>

Reduce usage of magic lengths of the supplied buffer by replacing them with the corresponding sizeof():s.

---
 drivers/iio/magnetometer/ak8975.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)
drivers/iio/magnetometer/ak8975.c +9 -6
index f58cbafb491a5e039b22fc928a6c58dd6ec0f297..fd42e20cd145860bdd1b143409de24b7145f8252 100644
@@ -490,8 +490,10 @@ static int ak8975_who_i_am(struct i2c_client *client,
 	 * AK8975   |  DEVICE_ID |  NA
 	 * AK8963   |  DEVICE_ID |  NA
 	 */
-	ret = i2c_smbus_read_i2c_block_data_or_emulated(
-			client, AK09912_REG_WIA1, 2, wia_val);
+	ret = i2c_smbus_read_i2c_block_data_or_emulated(client,
+							AK09912_REG_WIA1,
+							sizeof(wia_val),
+							wia_val);
 	if (ret < 0) {
 		dev_err(&client->dev, "Error reading WIA\n");
 		return ret;
@@ -610,9 +612,10 @@ static int ak8975_setup(struct i2c_client *client)
 	}
 
 	/* Get asa data and store in the device data. */
-	ret = i2c_smbus_read_i2c_block_data_or_emulated(
-			client, data->def->ctrl_regs[ASA_BASE],
-			3, data->asa);
+	ret = i2c_smbus_read_i2c_block_data_or_emulated(client,
+							data->def->ctrl_regs[ASA_BASE],
+							sizeof(data->asa),
+							data->asa);
 	if (ret < 0) {
 		dev_err(&client->dev, "Not able to read asa data\n");
 		return ret;
@@ -865,7 +868,7 @@ static void ak8975_fill_buffer(struct iio_dev *indio_dev)
 	 */
 	ret = i2c_smbus_read_i2c_block_data_or_emulated(client,
 							def->data_regs[0],
-							3 * sizeof(fval[0]),
+							sizeof(fval),
 							(u8 *)fval);
 	if (ret < 0)
 		goto unlock;
Signature · 1 line
2.47.3
#57 lore raw

PATCH v5 13/18 iio: magnetometer: ak8975: unify return code variable name

In reply to Joshua Crofts

From: Andy Shevchenko <[email protected]>

In one case 'rc' is used in the other 'err', the most use 'ret'. Make the latter use the former, id est 'ret'.

While at it, drop unneeded ' < 0' checks.

---
 drivers/iio/magnetometer/ak8975.c | 46 +++++++++++++++++++--------------------
 1 file changed, 23 insertions(+), 23 deletions(-)
drivers/iio/magnetometer/ak8975.c +23 -23
index fd42e20cd145860bdd1b143409de24b7145f8252..1c05e380a4d4f968f505a7d5c0acbec86ad949e1 100644
@@ -573,8 +573,8 @@ static irqreturn_t ak8975_irq_handler(int irq, void *data)
 static int ak8975_setup_irq(struct ak8975_data *data)
 {
 	struct i2c_client *client = data->client;
-	int rc;
 	int irq;
+	int ret;
 
 	init_waitqueue_head(&data->data_ready_queue);
 	clear_bit(0, &data->flags);
@@ -583,15 +583,15 @@ static int ak8975_setup_irq(struct ak8975_data *data)
 	else
 		irq = gpiod_to_irq(data->eoc_gpiod);
 
-	rc = devm_request_irq(&client->dev, irq, ak8975_irq_handler,
-			      IRQF_TRIGGER_RISING,
-			      dev_name(&client->dev), data);
-	if (rc < 0)
-		return rc;
+	ret = devm_request_irq(&client->dev, irq, ak8975_irq_handler,
+			       IRQF_TRIGGER_RISING,
+			       dev_name(&client->dev), data);
+	if (ret)
+		return ret;
 
 	data->eoc_irq = irq;
 
-	return rc;
+	return 0;
 }
 
 /*
@@ -907,8 +907,8 @@ static int ak8975_probe(struct i2c_client *client)
 	struct iio_dev *indio_dev;
 	struct gpio_desc *eoc_gpiod;
 	struct gpio_desc *reset_gpiod;
-	int err;
 	const char *name = NULL;
+	int ret;
 
 	/*
 	 * Grab and set up the supplied GPIO.
@@ -943,9 +943,9 @@ static int ak8975_probe(struct i2c_client *client)
 	data->reset_gpiod = reset_gpiod;
 	data->eoc_irq = 0;
 
-	err = iio_read_mount_matrix(&client->dev, &data->orientation);
-	if (err)
-		return err;
+	ret = iio_read_mount_matrix(&client->dev, &data->orientation);
+	if (ret)
+		return ret;
 
 	/* id will be NULL when enumerated via ACPI */
 	data->def = i2c_get_match_data(client);
@@ -966,20 +966,20 @@ static int ak8975_probe(struct i2c_client *client)
 	if (IS_ERR(data->vid))
 		return PTR_ERR(data->vid);
 
-	err = ak8975_power_on(data);
-	if (err)
-		return err;
+	ret = ak8975_power_on(data);
+	if (ret)
+		return ret;
 
-	err = ak8975_who_i_am(client, data->def->type);
-	if (err < 0) {
+	ret = ak8975_who_i_am(client, data->def->type);
+	if (ret) {
 		dev_err(&client->dev, "Unexpected device\n");
 		goto power_off;
 	}
 	dev_dbg(&client->dev, "Asahi compass chip %s\n", name);
 
 	/* Perform some basic start-of-day setup of the device. */
-	err = ak8975_setup(client);
-	if (err < 0) {
+	ret = ak8975_setup(client);
+	if (ret) {
 		dev_err(&client->dev, "%s initialization fails\n", name);
 		goto power_off;
 	}
@@ -992,15 +992,15 @@ static int ak8975_probe(struct i2c_client *client)
 	indio_dev->modes = INDIO_DIRECT_MODE;
 	indio_dev->name = name;
 
-	err = iio_triggered_buffer_setup(indio_dev, NULL, ak8975_handle_trigger,
+	ret = iio_triggered_buffer_setup(indio_dev, NULL, ak8975_handle_trigger,
 					 NULL);
-	if (err) {
+	if (ret) {
 		dev_err(&client->dev, "triggered buffer setup failed\n");
 		goto power_off;
 	}
 
-	err = iio_device_register(indio_dev);
-	if (err) {
+	ret = iio_device_register(indio_dev);
+	if (ret) {
 		dev_err(&client->dev, "device register failed\n");
 		goto cleanup_buffer;
 	}
@@ -1023,7 +1023,7 @@ static int ak8975_probe(struct i2c_client *client)
 	iio_triggered_buffer_cleanup(indio_dev);
 power_off:
 	ak8975_power_off(data);
-	return err;
+	return ret;
 }
 
 static void ak8975_remove(struct i2c_client *client)
Signature · 1 line
2.47.3
maintainer Applied
#58 lore raw

Re: [PATCH v5 13/18] iio: magnetometer: ak8975: unify return code variable name

Quoted 9 lines · On Tue, 05 May 2026 13:46:09 +0200 Joshua Crofts via B4 Relay…

From: Andy Shevchenko <[email protected]>

In one case 'rc' is used in the other 'err', the most use 'ret'. Make the latter use the former, id est 'ret'.

While at it, drop unneeded ' < 0' checks.

Applied.

#59 lore raw

PATCH v5 14/18 iio: magnetometer: ak8975: switch to using managed resources

In reply to Joshua Crofts

From: Andy Shevchenko <[email protected]>

Switch the driver to use managed resources (devm_*) which simplifier error handling and allows removing ak8975_remove() method from the driver.

Note, on error path we now also set mode to POWER_DOWN state which is fine. Even if the device is in that mode, there is no problem to set that mode again, it should be no-op.

---
 drivers/iio/magnetometer/ak8975.c | 77 ++++++++++++++++++++++-----------------
 1 file changed, 44 insertions(+), 33 deletions(-)
drivers/iio/magnetometer/ak8975.c +44 -33
index 1c05e380a4d4f968f505a7d5c0acbec86ad949e1..84ccbca758b0121a9c4930a368cb113471d389da 100644
@@ -900,9 +900,27 @@ static irqreturn_t ak8975_handle_trigger(int irq, void *p)
 	return IRQ_HANDLED;
 }
 
+static void devm_ak8975_power_off(void *data)
+{
+	struct ak8975_data *ak = data;
+	struct device *dev = &ak->client->dev;
+
+	/* Only power down if currently active */
+	if (pm_runtime_status_suspended(dev))
+		return;
+	/*
+	 * The device may not be runtime suspended when the driver is
+	 * removed, so we soft-stop the chip before hard-stopping the
+	 * regulators.
+	 */
+	ak8975_set_mode(data, POWER_DOWN);
+	ak8975_power_off(data);
+}
+
 static int ak8975_probe(struct i2c_client *client)
 {
 	const struct i2c_device_id *id = i2c_client_get_device_id(client);
+	struct device *dev = &client->dev;
 	struct ak8975_data *data;
 	struct iio_dev *indio_dev;
 	struct gpio_desc *eoc_gpiod;
@@ -970,10 +988,21 @@ static int ak8975_probe(struct i2c_client *client)
 	if (ret)
 		return ret;
 
+	/*
+	 * Set device as active early so pm_runtime_status_suspended() works
+	 * correctly if probe fails before pm_runtime is enabled. Do not attempt
+	 * to move this line lower.
+	 */
+	pm_runtime_set_active(dev);
+
+	ret = devm_add_action_or_reset(dev, devm_ak8975_power_off, data);
+	if (ret)
+		return ret;
+
 	ret = ak8975_who_i_am(client, data->def->type);
 	if (ret) {
 		dev_err(&client->dev, "Unexpected device\n");
-		goto power_off;
+		return ret;
 	}
 	dev_dbg(&client->dev, "Asahi compass chip %s\n", name);
 
@@ -981,10 +1010,13 @@ static int ak8975_probe(struct i2c_client *client)
 	ret = ak8975_setup(client);
 	if (ret) {
 		dev_err(&client->dev, "%s initialization fails\n", name);
-		goto power_off;
+		return ret;
 	}
 
-	mutex_init(&data->lock);
+	ret = devm_mutex_init(dev, &data->lock);
+	if (ret)
+		return ret;
+
 	indio_dev->channels = ak8975_channels;
 	indio_dev->num_channels = ARRAY_SIZE(ak8975_channels);
 	indio_dev->info = &ak8975_info;
@@ -992,52 +1024,32 @@ static int ak8975_probe(struct i2c_client *client)
 	indio_dev->modes = INDIO_DIRECT_MODE;
 	indio_dev->name = name;
 
-	ret = iio_triggered_buffer_setup(indio_dev, NULL, ak8975_handle_trigger,
-					 NULL);
+	ret = devm_iio_triggered_buffer_setup(dev, indio_dev, NULL,
+					      ak8975_handle_trigger, NULL);
 	if (ret) {
 		dev_err(&client->dev, "triggered buffer setup failed\n");
-		goto power_off;
+		return ret;
 	}
 
-	ret = iio_device_register(indio_dev);
+	ret = devm_iio_device_register(dev, indio_dev);
 	if (ret) {
 		dev_err(&client->dev, "device register failed\n");
-		goto cleanup_buffer;
+		return ret;
 	}
 
 	/* Enable runtime PM */
-	pm_runtime_get_noresume(&client->dev);
-	pm_runtime_set_active(&client->dev);
-	pm_runtime_enable(&client->dev);
+	ret = devm_pm_runtime_enable(dev);
+	if (ret)
+		return ret;
+
 	/*
 	 * The device comes online in 500us, so add two orders of magnitude
 	 * of delay before autosuspending: 50 ms.
 	 */
 	pm_runtime_set_autosuspend_delay(&client->dev, 50);
 	pm_runtime_use_autosuspend(&client->dev);
-	pm_runtime_put(&client->dev);
 
 	return 0;
-
-cleanup_buffer:
-	iio_triggered_buffer_cleanup(indio_dev);
-power_off:
-	ak8975_power_off(data);
-	return ret;
-}
-
-static void ak8975_remove(struct i2c_client *client)
-{
-	struct iio_dev *indio_dev = i2c_get_clientdata(client);
-	struct ak8975_data *data = iio_priv(indio_dev);
-
-	pm_runtime_get_sync(&client->dev);
-	pm_runtime_put_noidle(&client->dev);
-	pm_runtime_disable(&client->dev);
-	iio_device_unregister(indio_dev);
-	iio_triggered_buffer_cleanup(indio_dev);
-	ak8975_set_mode(data, POWER_DOWN);
-	ak8975_power_off(data);
 }
 
 static int ak8975_runtime_suspend(struct device *dev)
@@ -1131,7 +1143,6 @@ static struct i2c_driver ak8975_driver = {
 		.acpi_match_table = ak_acpi_match,
 	},
 	.probe		= ak8975_probe,
-	.remove		= ak8975_remove,
 	.id_table	= ak8975_id,
 };
 module_i2c_driver(ak8975_driver);
Signature · 1 line
2.47.3
#60 lore raw

Re: [PATCH v5 14/18] iio: magnetometer: ak8975: switch to using managed resources

Quoted 7 lines · On Tue, 5 May 2026 at 13:46, Joshua Crofts via B4 Relay…

From: Andy Shevchenko <[email protected]>

Switch the driver to use managed resources (devm_*) which simplifier error handling and allows removing ak8975_remove() method from the driver.

Quoted 7 lines

Note, on error path we now also set mode to POWER_DOWN state which is fine. Even if the device is in that mode, there is no problem to set that mode again, it should be no-op.

BTW, Sashiko timed out while reviewing this patch, didn't even know that's possible.

Signature · 3 lines
Kind regards

CJD
#61 lore raw

Re: [PATCH v5 14/18] iio: magnetometer: ak8975: switch to using managed resources

Quoted 11 lines · Joshua Crofts
Quoted 9 lines · On Tue, 5 May 2026 at 13:46, Joshua Crofts via B4 Relay…

Switch the driver to use managed resources (devm_*) which simplifier error handling and allows removing ak8975_remove() method from the driver.

Note, on error path we now also set mode to POWER_DOWN state which is fine. Even if the device is in that mode, there is no problem to set that mode again, it should be no-op.

Quoted 6 lines
Quoted 3 lines

BTW, Sashiko timed out while reviewing this patch, didn't even know that's possible.

You can ask Roman what to do in such a case.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
maintainer
#62 lore raw

Re: [PATCH v5 14/18] iio: magnetometer: ak8975: switch to using managed resources

In reply to Joshua Crofts
Quoted 6 lines · On Tue, 05 May 2026 13:46:10 +0200 Joshua Crofts via B4 Relay…

From: Andy Shevchenko <[email protected]>

Switch the driver to use managed resources (devm_*) which simplifier error handling and allows removing ak8975_remove() method from the driver.

Quoted 7 lines

Note, on error path we now also set mode to POWER_DOWN state which is fine. Even if the device is in that mode, there is no problem to set that mode again, it should be no-op.

Hi Joshua

Looks fine, but one stale comment needs an update.

Jonathan

Quoted 12 lines
---
 drivers/iio/magnetometer/ak8975.c | 77 ++++++++++++++++++++++-----------------
 1 file changed, 44 insertions(+), 33 deletions(-)
drivers/iio/magnetometer/ak8975.c +0 -0
index 1c05e380a4d4f968f505a7d5c0acbec86ad949e1..84ccbca758b0121a9c4930a368cb113471d389da 100644
@@ -900,9 +900,27 @@ static irqreturn_t ak8975_handle_trigger(int irq, void *p)
 	return IRQ_HANDLED;
 }
 

drivers/iio/magnetometer/ak8975.c static irqreturn_t ak8975_handle_trigger(int irq, void *p) · line 912

Quoted 10 lines
drivers/iio/magnetometer/ak8975.c +10 -0
+static void devm_ak8975_power_off(void *data)
+{
+	struct ak8975_data *ak = data;
+	struct device *dev = &ak->client->dev;
+
+	/* Only power down if currently active */
+	if (pm_runtime_status_suspended(dev))
+		return;
+	/*
+	 * The device may not be runtime suspended when the driver is

If we get here it's definitely not runtime suspended. Hence this comment needs a rewrite.

drivers/iio/magnetometer/ak8975.c static irqreturn_t ak8975_handle_trigger(int irq, void *p) · line 916

Quoted 4 lines
drivers/iio/magnetometer/ak8975.c +4 -0
+	 * removed, so we soft-stop the chip before hard-stopping the
+	 * regulators.
+	 */
+	ak8975_set_mode(data, POWER_DOWN);

I went looking into how we get into different modes...
This deals with a latent bug in ak8975_setup() in that it can be in fuse mode in some exit paths. Arguably if we had an i2c bus fail we probably can't change the mode anyway, but we should be trying from point of view of no side effects. I wonder if we should fix that in a precursor to this patch (given the amount of churn from earlier patches an obscure corner case of the bug - I don't plan to mark it for backports either way!)

Maybe it's just not worth the bother given this is now here anyway.

Quoted 63 lines
drivers/iio/magnetometer/ak8975.c +24 -6
+	ak8975_power_off(data);
+}
+
 static int ak8975_probe(struct i2c_client *client)
 {
 	const struct i2c_device_id *id = i2c_client_get_device_id(client);
+	struct device *dev = &client->dev;
 	struct ak8975_data *data;
 	struct iio_dev *indio_dev;
 	struct gpio_desc *eoc_gpiod;
@@ -970,10 +988,21 @@ static int ak8975_probe(struct i2c_client *client)
 	if (ret)
 		return ret;
 
+	/*
+	 * Set device as active early so pm_runtime_status_suspended() works
+	 * correctly if probe fails before pm_runtime is enabled. Do not attempt
+	 * to move this line lower.
+	 */
+	pm_runtime_set_active(dev);
+
+	ret = devm_add_action_or_reset(dev, devm_ak8975_power_off, data);
+	if (ret)
+		return ret;
+
 	ret = ak8975_who_i_am(client, data->def->type);
 	if (ret) {
 		dev_err(&client->dev, "Unexpected device\n");
-		goto power_off;
+		return ret;
 	}
 	dev_dbg(&client->dev, "Asahi compass chip %s\n", name);
 
@@ -981,10 +1010,13 @@ static int ak8975_probe(struct i2c_client *client)
 	ret = ak8975_setup(client);
 	if (ret) {
 		dev_err(&client->dev, "%s initialization fails\n", name);
-		goto power_off;
+		return ret;
 	}
 
-	mutex_init(&data->lock);
+	ret = devm_mutex_init(dev, &data->lock);
+	if (ret)
+		return ret;
+
 	indio_dev->channels = ak8975_channels;
 	indio_dev->num_channels = ARRAY_SIZE(ak8975_channels);
 	indio_dev->info = &ak8975_info;
@@ -992,52 +1024,32 @@ static int ak8975_probe(struct i2c_client *client)
 	indio_dev->modes = INDIO_DIRECT_MODE;
 	indio_dev->name = name;
 
-	ret = iio_triggered_buffer_setup(indio_dev, NULL, ak8975_handle_trigger,
-					 NULL);
+	ret = devm_iio_triggered_buffer_setup(dev, indio_dev, NULL,
+					      ak8975_handle_trigger, NULL);
 	if (ret) {
 		dev_err(&client->dev, "triggered buffer setup failed\n");
-		goto power_off;
+		return ret;
 	}
 

drivers/iio/magnetometer/ak8975.c static int ak8975_probe(struct i2c_client *client) · line 1009

Quoted 10 lines
drivers/iio/magnetometer/ak8975.c +2 -3
-	ret = iio_device_register(indio_dev);
+	ret = devm_iio_device_register(dev, indio_dev);
 	if (ret) {
 		dev_err(&client->dev, "device register failed\n");
-		goto cleanup_buffer;
+		return ret;
 	}
 
 	/* Enable runtime PM */
-	pm_runtime_get_noresume(&client->dev);

huh. Not sure what this was doing in original code.
I guess forced an immediate power down rather that after an autosuspend period. I think that's fine to change - however perhaps do a bit of analysis on whether my assumption is correct on what this did and then mention it in the commit message.

Quoted 16 lines
drivers/iio/magnetometer/ak8975.c +4 -3
-	pm_runtime_set_active(&client->dev);
 
-	pm_runtime_enable(&client->dev);
+	ret = devm_pm_runtime_enable(dev);
+	if (ret)
+		return ret;
+
 	/*
 	 * The device comes online in 500us, so add two orders of magnitude
 	 * of delay before autosuspending: 50 ms.
 	 */
 	pm_runtime_set_autosuspend_delay(&client->dev, 50);
 	pm_runtime_use_autosuspend(&client->dev);
-	pm_runtime_put(&client->dev);
 
 	return 0;
#63 lore raw

Re: [PATCH v5 14/18] iio: magnetometer: ak8975: switch to using managed resources

Quoted 35 lines · Jonathan Cameron
Quoted 6 lines · On Tue, 05 May 2026 13:46:10 +0200 Joshua Crofts via B4 Relay…

From: Andy Shevchenko <[email protected]>

Switch the driver to use managed resources (devm_*) which simplifier error handling and allows removing ak8975_remove() method from the driver.

Quoted 7 lines

Note, on error path we now also set mode to POWER_DOWN state which is fine. Even if the device is in that mode, there is no problem to set that mode again, it should be no-op.

Hi Joshua

Looks fine, but one stale comment needs an update.

Jonathan

Quoted 12 lines
---
 drivers/iio/magnetometer/ak8975.c | 77 ++++++++++++++++++++++-----------------
 1 file changed, 44 insertions(+), 33 deletions(-)
drivers/iio/magnetometer/ak8975.c +0 -0
index 1c05e380a4d4f968f505a7d5c0acbec86ad949e1..84ccbca758b0121a9c4930a368cb113471d389da 100644
@@ -900,9 +900,27 @@ static irqreturn_t ak8975_handle_trigger(int irq, void *p)
      return IRQ_HANDLED;
 }
 
Quoted 13 lines

drivers/iio/magnetometer/ak8975.c static irqreturn_t ak8975_handle_trigger(int irq, void *p) · line 912

Quoted 10 lines
drivers/iio/magnetometer/ak8975.c +10 -0
+static void devm_ak8975_power_off(void *data)
+{
+     struct ak8975_data *ak = data;
+     struct device *dev = &ak->client->dev;
+
+     /* Only power down if currently active */
+     if (pm_runtime_status_suspended(dev))
+             return;
+     /*
+      * The device may not be runtime suspended when the driver is

If we get here it's definitely not runtime suspended. Hence this comment needs a rewrite.

Missed that one, thanks.

Quoted 85 lines

drivers/iio/magnetometer/ak8975.c static irqreturn_t ak8975_handle_trigger(int irq, void *p) · line 916

Quoted 4 lines
drivers/iio/magnetometer/ak8975.c +4 -0
+      * removed, so we soft-stop the chip before hard-stopping the
+      * regulators.
+      */
+     ak8975_set_mode(data, POWER_DOWN);

I went looking into how we get into different modes...
This deals with a latent bug in ak8975_setup() in that it can be in fuse mode in some exit paths. Arguably if we had an i2c bus fail we probably can't change the mode anyway, but we should be trying from point of view of no side effects. I wonder if we should fix that in a precursor to this patch (given the amount of churn from earlier patches an obscure corner case of the bug - I don't plan to mark it for backports either way!)

Maybe it's just not worth the bother given this is now here anyway.

Quoted 71 lines
drivers/iio/magnetometer/ak8975.c +26 -8
+     ak8975_power_off(data);
+}
+
 static int ak8975_probe(struct i2c_client *client)
 {
      const struct i2c_device_id *id = i2c_client_get_device_id(client);
+     struct device *dev = &client->dev;
      struct ak8975_data *data;
      struct iio_dev *indio_dev;
      struct gpio_desc *eoc_gpiod;
@@ -970,10 +988,21 @@ static int ak8975_probe(struct i2c_client *client)
      if (ret)
              return ret;
 
+     /*
+      * Set device as active early so pm_runtime_status_suspended() works
+      * correctly if probe fails before pm_runtime is enabled. Do not attempt
+      * to move this line lower.
+      */
+     pm_runtime_set_active(dev);
+
+     ret = devm_add_action_or_reset(dev, devm_ak8975_power_off, data);
+     if (ret)
+             return ret;
+
      ret = ak8975_who_i_am(client, data->def->type);
      if (ret) {
              dev_err(&client->dev, "Unexpected device\n");
-             goto power_off;
+             return ret;
      }
      dev_dbg(&client->dev, "Asahi compass chip %s\n", name);
 
@@ -981,10 +1010,13 @@ static int ak8975_probe(struct i2c_client *client)
      ret = ak8975_setup(client);
      if (ret) {
              dev_err(&client->dev, "%s initialization fails\n", name);
-             goto power_off;
+             return ret;
      }
 
-     mutex_init(&data->lock);
+     ret = devm_mutex_init(dev, &data->lock);
+     if (ret)
+             return ret;
+
      indio_dev->channels = ak8975_channels;
      indio_dev->num_channels = ARRAY_SIZE(ak8975_channels);
      indio_dev->info = &ak8975_info;
@@ -992,52 +1024,32 @@ static int ak8975_probe(struct i2c_client *client)
      indio_dev->modes = INDIO_DIRECT_MODE;
      indio_dev->name = name;
 
-     ret = iio_triggered_buffer_setup(indio_dev, NULL, ak8975_handle_trigger,
-                                      NULL);
+     ret = devm_iio_triggered_buffer_setup(dev, indio_dev, NULL,
+                                           ak8975_handle_trigger, NULL);
      if (ret) {
              dev_err(&client->dev, "triggered buffer setup failed\n");
-             goto power_off;
+             return ret;
      }
 
-     ret = iio_device_register(indio_dev);
+     ret = devm_iio_device_register(dev, indio_dev);
      if (ret) {
              dev_err(&client->dev, "device register failed\n");
-             goto cleanup_buffer;
+             return ret;
      }
 
Quoted 8 lines

drivers/iio/magnetometer/ak8975.c static int ak8975_probe(struct i2c_client *client) · line 1009

Quoted 2 lines
drivers/iio/magnetometer/ak8975.c +0 -1
      /* Enable runtime PM */
-     pm_runtime_get_noresume(&client->dev);

huh. Not sure what this was doing in original code.
I guess forced an immediate power down rather that after an autosuspend period. I think that's fine to change - however perhaps do a bit of analysis on whether my assumption is correct on what this did and then mention it in the commit message.

This is a fun one. pm_runtime_get_noresume() increments the counter, then later on pm_runtime_put() was called, essentially tricking pm_runtime into thinking that some task has ended, forcing it to autosuspend. Looking at the code now, the device won't autosuspend after probe() is finished, it will just be active until the driver is accessed from userspace (and subsequently completes the task).

Sashiko is pretty impressive, it took me a long time to see what's going on.

Quoted 14 lines
Quoted 4 lines
drivers/iio/magnetometer/ak8975.c +1 -2
-     pm_runtime_set_active(&client->dev);
 
-     pm_runtime_enable(&client->dev);
+     ret = devm_pm_runtime_enable(dev);

drivers/iio/magnetometer/ak8975.c static int ak8975_probe(struct i2c_client *client) · line 1018

Quoted 10 lines
drivers/iio/magnetometer/ak8975.c +3 -1
+     if (ret)
+             return ret;
+
      /*
       * The device comes online in 500us, so add two orders of magnitude
       * of delay before autosuspending: 50 ms.
       */
      pm_runtime_set_autosuspend_delay(&client->dev, 50);
      pm_runtime_use_autosuspend(&client->dev);
-     pm_runtime_put(&client->dev);

I'd just add pm_runtime_mark_last_busy() and pm_request_autosuspend() here to make the device suspend.

Perhaps unrelated, however I found an issue with the pm_runtime documentation, as it incorrectly says that pm_runtime_mark_last_busy() is always called with an autosuspend function (which is not the case, looking at the definition).

"Drivers may also continue to use the non-autosuspend helper functions; they will
behave normally, which means sometimes taking the autosuspend delay into account (see pm_runtime_idle). The autosuspend variants of the functions also call
pm_runtime_mark_last_busy()."

Signature · 3 lines
Kind regards

CJD
#64 lore raw

Re: [PATCH v5 14/18] iio: magnetometer: ak8975: switch to using managed resources

Quoted 3 lines · Joshua Crofts
Quoted 2 lines · Jonathan Cameron

On Tue, 05 May 2026 13:46:10 +0200
Joshua Crofts via B4 Relay <[email protected]> wrote:

...

Quoted 2 lines

I'd just add pm_runtime_mark_last_busy() and pm_request_autosuspend() here to make the device suspend.

pm_runtime_put_autosuspend() is exactly doing that in one call. That's what documentation says about, and that's correct, however it might be not so clear.

Quoted 10 lines

Perhaps unrelated, however I found an issue with the pm_runtime documentation, as it incorrectly says that pm_runtime_mark_last_busy() is always called with an autosuspend function (which is not the case, looking at the definition).

"Drivers may also continue to use the non-autosuspend helper functions; they will
behave normally, which means sometimes taking the autosuspend delay into account (see pm_runtime_idle). The autosuspend variants of the functions also call
pm_runtime_mark_last_busy()."

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#65 lore raw

Re: [PATCH v5 14/18] iio: magnetometer: ak8975: switch to using managed resources

Quoted 6 lines · Andy Shevchenko
Quoted 3 lines · Joshua Crofts
Quoted 2 lines · Jonathan Cameron

On Tue, 05 May 2026 13:46:10 +0200
Joshua Crofts via B4 Relay <[email protected]> wrote:

Quoted 7 lines

...

Quoted 2 lines

I'd just add pm_runtime_mark_last_busy() and pm_request_autosuspend() here to make the device suspend.

pm_runtime_put_autosuspend() is exactly doing that in one call. That's what documentation says about, and that's correct, however it might be not so clear.

Maybe I didn't look enough, but I couldn't find a definition of pm_runtime_put_autosuspend()
that actually calls mark_last_busy().

Either way, it wouldn't hurt to add it manually, just to be sure?

Signature · 3 lines
Kind regards

CJD
#66 lore raw

Re: [PATCH v5 14/18] iio: magnetometer: ak8975: switch to using managed resources

Quoted 6 lines · Joshua Crofts
Quoted 4 lines · Andy Shevchenko
Quoted 3 lines · Joshua Crofts
Quoted 2 lines · Jonathan Cameron

On Tue, 05 May 2026 13:46:10 +0200
Joshua Crofts via B4 Relay <[email protected]> wrote:

...

Quoted 4 lines
Quoted 4 lines
Quoted 2 lines

I'd just add pm_runtime_mark_last_busy() and pm_request_autosuspend() here to make the device suspend.

pm_runtime_put_autosuspend() is exactly doing that in one call. That's what

(Okay, this should be read as "pm_request_autosuspend()".)

Quoted 5 lines
Quoted 1 line

documentation says about, and that's correct, however it might be not so clear.

Maybe I didn't look enough, but I couldn't find a definition of pm_runtime_put_autosuspend()
that actually calls mark_last_busy().

Hmm... It's just there.

static inline int pm_runtime_put_autosuspend(struct device *dev)

{
        pm_runtime_mark_last_busy(dev);
        return __pm_runtime_put_autosuspend(dev);
}

Even in pm_request_autosuspend().

Quoted 1 line

Either way, it wouldn't hurt to add it manually, just to be sure?

Why? the whole point that it's not needed when autosuspend is in use.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#67 lore raw

Re: [PATCH v5 14/18] iio: magnetometer: ak8975: switch to using managed resources

Quoted 26 lines · Andy Shevchenko
Quoted 6 lines · Joshua Crofts
Quoted 4 lines · Andy Shevchenko
Quoted 3 lines · Joshua Crofts
Quoted 2 lines · Jonathan Cameron

On Tue, 05 May 2026 13:46:10 +0200
Joshua Crofts via B4 Relay <[email protected]> wrote:

...

Quoted 4 lines
Quoted 4 lines
Quoted 2 lines

I'd just add pm_runtime_mark_last_busy() and pm_request_autosuspend() here to make the device suspend.

pm_runtime_put_autosuspend() is exactly doing that in one call. That's what

(Okay, this should be read as "pm_request_autosuspend()".)

Quoted 5 lines
Quoted 1 line

documentation says about, and that's correct, however it might be not so clear.

Maybe I didn't look enough, but I couldn't find a definition of pm_runtime_put_autosuspend()
that actually calls mark_last_busy().

Hmm... It's just there.

Quoted 7 lines

static inline int pm_runtime_put_autosuspend(struct device *dev)

{
        pm_runtime_mark_last_busy(dev);
        return __pm_runtime_put_autosuspend(dev);
}

Even in pm_request_autosuspend().

How strange. I guess no changes then.

Thanks for the help!

Signature · 3 lines
Kind regards

CJD
#68 lore raw

PATCH v5 15/18 iio: magnetometer: ak8975: consistently use 'data' parameter

In reply to Joshua Crofts

From: Andy Shevchenko <[email protected]>

Some of the functions use 'client', some use 'data', and some use both. Refactor the driver to consistently use 'data' in all cases.

---
 drivers/iio/magnetometer/ak8975.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
drivers/iio/magnetometer/ak8975.c +6 -6
index 84ccbca758b0121a9c4930a368cb113471d389da..d40ce26e3a47d5a5c9c81d49c128d136208a343a 100644
@@ -474,9 +474,10 @@ static void ak8975_power_off(const struct ak8975_data *data)
  * Return 0 if the i2c device is the one we expect.
  * return a negative error number otherwise
  */
-static int ak8975_who_i_am(struct i2c_client *client,
+static int ak8975_who_i_am(const struct ak8975_data *data,
 			   enum asahi_compass_chipset type)
 {
+	struct i2c_client *client = data->client;
 	u8 wia_val[2];
 	int ret;
 
@@ -598,10 +599,9 @@ static int ak8975_setup_irq(struct ak8975_data *data)
  * Perform some start-of-day setup, including reading the asa calibration
  * values and caching them.
  */
-static int ak8975_setup(struct i2c_client *client)
+static int ak8975_setup(struct ak8975_data *data)
 {
-	struct iio_dev *indio_dev = i2c_get_clientdata(client);
-	struct ak8975_data *data = iio_priv(indio_dev);
+	struct i2c_client *client = data->client;
 	int ret;
 
 	/* Write the fused rom access mode. */
@@ -999,7 +999,7 @@ static int ak8975_probe(struct i2c_client *client)
 	if (ret)
 		return ret;
 
-	ret = ak8975_who_i_am(client, data->def->type);
+	ret = ak8975_who_i_am(data, data->def->type);
 	if (ret) {
 		dev_err(&client->dev, "Unexpected device\n");
 		return ret;
@@ -1007,7 +1007,7 @@ static int ak8975_probe(struct i2c_client *client)
 	dev_dbg(&client->dev, "Asahi compass chip %s\n", name);
 
 	/* Perform some basic start-of-day setup of the device. */
-	ret = ak8975_setup(client);
+	ret = ak8975_setup(data);
 	if (ret) {
 		dev_err(&client->dev, "%s initialization fails\n", name);
 		return ret;
Signature · 1 line
2.47.3
#69 lore raw

PATCH v5 16/18 iio: magnetometer: ak8975: unify messages with help of dev_err_probe()

In reply to Joshua Crofts

From: Andy Shevchenko <[email protected]>

Unify error messages that might appear during probe phase by switching to use dev_err_probe().

---
 drivers/iio/magnetometer/ak8975.c | 30 ++++++++++--------------------
 1 file changed, 10 insertions(+), 20 deletions(-)
drivers/iio/magnetometer/ak8975.c +10 -20
index d40ce26e3a47d5a5c9c81d49c128d136208a343a..b07c56312f7df07ea21c3f378b5721568cb66454 100644
@@ -495,10 +495,8 @@ static int ak8975_who_i_am(const struct ak8975_data *data,
 							AK09912_REG_WIA1,
 							sizeof(wia_val),
 							wia_val);
-	if (ret < 0) {
-		dev_err(&client->dev, "Error reading WIA\n");
-		return ret;
-	}
+	if (ret < 0)
+		return dev_err_probe(&client->dev, ret, "Error reading WIA\n");
 
 	if (wia_val[0] != AK8975_DEVICE_ID)
 		return -ENODEV;
@@ -1000,18 +998,14 @@ static int ak8975_probe(struct i2c_client *client)
 		return ret;
 
 	ret = ak8975_who_i_am(data, data->def->type);
-	if (ret) {
-		dev_err(&client->dev, "Unexpected device\n");
-		return ret;
-	}
+	if (ret)
+		return dev_err_probe(dev, ret, "Unexpected device\n");
 	dev_dbg(&client->dev, "Asahi compass chip %s\n", name);
 
 	/* Perform some basic start-of-day setup of the device. */
 	ret = ak8975_setup(data);
-	if (ret) {
-		dev_err(&client->dev, "%s initialization fails\n", name);
-		return ret;
-	}
+	if (ret)
+		return dev_err_probe(dev, ret, "%s initialization fails\n", name);
 
 	ret = devm_mutex_init(dev, &data->lock);
 	if (ret)
@@ -1026,16 +1020,12 @@ static int ak8975_probe(struct i2c_client *client)
 
 	ret = devm_iio_triggered_buffer_setup(dev, indio_dev, NULL,
 					      ak8975_handle_trigger, NULL);
-	if (ret) {
-		dev_err(&client->dev, "triggered buffer setup failed\n");
-		return ret;
-	}
+	if (ret)
+		return dev_err_probe(dev, ret, "triggered buffer setup failed\n");
 
 	ret = devm_iio_device_register(dev, indio_dev);
-	if (ret) {
-		dev_err(&client->dev, "device register failed\n");
-		return ret;
-	}
+	if (ret)
+		return dev_err_probe(dev, ret, "device register failed\n");
 
 	/* Enable runtime PM */
 	ret = devm_pm_runtime_enable(dev);
Signature · 1 line
2.47.3
#70 lore raw

PATCH v5 17/18 iio: magnetometer: ak8975: use temporary variable for struct device

In reply to Joshua Crofts

From: Andy Shevchenko <[email protected]>

Use temporary variable for struct device to make code neater.

---
 drivers/iio/magnetometer/ak8975.c | 61 +++++++++++++++++++--------------------
 1 file changed, 30 insertions(+), 31 deletions(-)
drivers/iio/magnetometer/ak8975.c +30 -31
index b07c56312f7df07ea21c3f378b5721568cb66454..1c96df3e1ad95d268470609c61be8ee4a709675d 100644
@@ -433,18 +433,17 @@ struct ak8975_data {
 /* Enable attached power regulator if any. */
 static int ak8975_power_on(const struct ak8975_data *data)
 {
+	struct device *dev = &data->client->dev;
 	int ret;
 
 	ret = regulator_enable(data->vdd);
 	if (ret) {
-		dev_warn(&data->client->dev,
-			 "Failed to enable specified Vdd supply\n");
+		dev_warn(dev, "Failed to enable specified Vdd supply\n");
 		return ret;
 	}
 	ret = regulator_enable(data->vid);
 	if (ret) {
-		dev_warn(&data->client->dev,
-			 "Failed to enable specified Vid supply\n");
+		dev_warn(dev, "Failed to enable specified Vid supply\n");
 		regulator_disable(data->vdd);
 		return ret;
 	}
@@ -572,6 +571,7 @@ static irqreturn_t ak8975_irq_handler(int irq, void *data)
 static int ak8975_setup_irq(struct ak8975_data *data)
 {
 	struct i2c_client *client = data->client;
+	struct device *dev = &client->dev;
 	int irq;
 	int ret;
 
@@ -582,9 +582,8 @@ static int ak8975_setup_irq(struct ak8975_data *data)
 	else
 		irq = gpiod_to_irq(data->eoc_gpiod);
 
-	ret = devm_request_irq(&client->dev, irq, ak8975_irq_handler,
-			       IRQF_TRIGGER_RISING,
-			       dev_name(&client->dev), data);
+	ret = devm_request_irq(dev, irq, ak8975_irq_handler, IRQF_TRIGGER_RISING,
+			       dev_name(dev), data);
 	if (ret)
 		return ret;
 
@@ -600,12 +599,13 @@ static int ak8975_setup_irq(struct ak8975_data *data)
 static int ak8975_setup(struct ak8975_data *data)
 {
 	struct i2c_client *client = data->client;
+	struct device *dev = &client->dev;
 	int ret;
 
 	/* Write the fused rom access mode. */
 	ret = ak8975_set_mode(data, FUSE_ROM);
 	if (ret < 0) {
-		dev_err(&client->dev, "Error in setting fuse access mode\n");
+		dev_err(dev, "Error in setting fuse access mode\n");
 		return ret;
 	}
 
@@ -615,22 +615,21 @@ static int ak8975_setup(struct ak8975_data *data)
 							sizeof(data->asa),
 							data->asa);
 	if (ret < 0) {
-		dev_err(&client->dev, "Not able to read asa data\n");
+		dev_err(dev, "Not able to read asa data\n");
 		return ret;
 	}
 
 	/* After reading fuse ROM data set power-down mode */
 	ret = ak8975_set_mode(data, POWER_DOWN);
 	if (ret < 0) {
-		dev_err(&client->dev, "Error in setting power-down mode\n");
+		dev_err(dev, "Error in setting power-down mode\n");
 		return ret;
 	}
 
 	if (data->eoc_gpiod || client->irq > 0) {
 		ret = ak8975_setup_irq(data);
 		if (ret < 0) {
-			dev_err(&client->dev,
-				"Error setting data ready interrupt\n");
+			dev_err(dev, "Error setting data ready interrupt\n");
 			return ret;
 		}
 	}
@@ -738,10 +737,11 @@ static int ak8975_read_axis(struct iio_dev *indio_dev, int index, int *val)
 	struct ak8975_data *data = iio_priv(indio_dev);
 	const struct i2c_client *client = data->client;
 	const struct ak_def *def = data->def;
+	struct device *dev = &data->client->dev;
 	__le16 rval;
 	int ret;
 
-	pm_runtime_get_sync(&data->client->dev);
+	pm_runtime_get_sync(dev);
 
 	mutex_lock(&data->lock);
 
@@ -759,20 +759,20 @@ static int ak8975_read_axis(struct iio_dev *indio_dev, int index, int *val)
 	/* Read out ST2 for release lock on measurement data. */
 	ret = i2c_smbus_read_byte_data(client, data->def->ctrl_regs[ST2]);
 	if (ret < 0) {
-		dev_err(&client->dev, "Error in reading ST2\n");
+		dev_err(dev, "Error in reading ST2\n");
 		goto exit;
 	}
 
 	if (ret & (data->def->ctrl_masks[ST2_DERR] |
 		   data->def->ctrl_masks[ST2_HOFL])) {
-		dev_err(&client->dev, "ST2 status error 0x%x\n", ret);
+		dev_err(dev, "ST2 status error 0x%x\n", ret);
 		ret = -EINVAL;
 		goto exit;
 	}
 
 	mutex_unlock(&data->lock);
 
-	pm_runtime_put_autosuspend(&data->client->dev);
+	pm_runtime_put_autosuspend(dev);
 
 	/* Swap bytes and convert to valid range. */
 	*val = clamp_t(s16, le16_to_cpu(rval), -def->range, def->range);
@@ -781,7 +781,7 @@ static int ak8975_read_axis(struct iio_dev *indio_dev, int index, int *val)
 
 exit:
 	mutex_unlock(&data->lock);
-	dev_err(&client->dev, "Error in reading axis\n");
+	dev_err(dev, "Error in reading axis\n");
 	return ret;
 }
 
@@ -931,7 +931,7 @@ static int ak8975_probe(struct i2c_client *client)
 	 * We may not have a GPIO based IRQ to scan, that is fine, we will
 	 * poll if so.
 	 */
-	eoc_gpiod = devm_gpiod_get_optional(&client->dev, NULL, GPIOD_IN);
+	eoc_gpiod = devm_gpiod_get_optional(dev, NULL, GPIOD_IN);
 	if (IS_ERR(eoc_gpiod))
 		return PTR_ERR(eoc_gpiod);
 	gpiod_set_consumer_name(eoc_gpiod, "ak_8975");
@@ -941,13 +941,12 @@ static int ak8975_probe(struct i2c_client *client)
 	 * deassert reset on ak8975_power_on() and assert reset on
 	 * ak8975_power_off().
 	 */
-	reset_gpiod = devm_gpiod_get_optional(&client->dev,
-					      "reset", GPIOD_OUT_HIGH);
+	reset_gpiod = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH);
 	if (IS_ERR(reset_gpiod))
 		return PTR_ERR(reset_gpiod);
 
 	/* Register with IIO */
-	indio_dev = devm_iio_device_alloc(&client->dev, sizeof(*data));
+	indio_dev = devm_iio_device_alloc(dev, sizeof(*data));
 	if (indio_dev == NULL)
 		return -ENOMEM;
 
@@ -959,7 +958,7 @@ static int ak8975_probe(struct i2c_client *client)
 	data->reset_gpiod = reset_gpiod;
 	data->eoc_irq = 0;
 
-	ret = iio_read_mount_matrix(&client->dev, &data->orientation);
+	ret = iio_read_mount_matrix(dev, &data->orientation);
 	if (ret)
 		return ret;
 
@@ -969,16 +968,16 @@ static int ak8975_probe(struct i2c_client *client)
 		return -ENODEV;
 
 	/* If enumerated via firmware node, fix the ABI */
-	if (dev_fwnode(&client->dev))
-		name = dev_name(&client->dev);
+	if (dev_fwnode(dev))
+		name = dev_name(dev);
 	else
 		name = id->name;
 
 	/* Fetch the regulators */
-	data->vdd = devm_regulator_get(&client->dev, "vdd");
+	data->vdd = devm_regulator_get(dev, "vdd");
 	if (IS_ERR(data->vdd))
 		return PTR_ERR(data->vdd);
-	data->vid = devm_regulator_get(&client->dev, "vid");
+	data->vid = devm_regulator_get(dev, "vid");
 	if (IS_ERR(data->vid))
 		return PTR_ERR(data->vid);
 
@@ -1000,7 +999,7 @@ static int ak8975_probe(struct i2c_client *client)
 	ret = ak8975_who_i_am(data, data->def->type);
 	if (ret)
 		return dev_err_probe(dev, ret, "Unexpected device\n");
-	dev_dbg(&client->dev, "Asahi compass chip %s\n", name);
+	dev_dbg(dev, "Asahi compass chip %s\n", name);
 
 	/* Perform some basic start-of-day setup of the device. */
 	ret = ak8975_setup(data);
@@ -1036,8 +1035,8 @@ static int ak8975_probe(struct i2c_client *client)
 	 * The device comes online in 500us, so add two orders of magnitude
 	 * of delay before autosuspending: 50 ms.
 	 */
-	pm_runtime_set_autosuspend_delay(&client->dev, 50);
-	pm_runtime_use_autosuspend(&client->dev);
+	pm_runtime_set_autosuspend_delay(dev, 50);
+	pm_runtime_use_autosuspend(dev);
 
 	return 0;
 }
@@ -1052,7 +1051,7 @@ static int ak8975_runtime_suspend(struct device *dev)
 	/* Set the device in power down if it wasn't already */
 	ret = ak8975_set_mode(data, POWER_DOWN);
 	if (ret < 0) {
-		dev_err(&client->dev, "Error in setting power-down mode\n");
+		dev_err(dev, "Error in setting power-down mode\n");
 		return ret;
 	}
 	/* Next cut the regulators */
@@ -1076,7 +1075,7 @@ static int ak8975_runtime_resume(struct device *dev)
 	 */
 	ret = ak8975_set_mode(data, POWER_DOWN);
 	if (ret < 0) {
-		dev_err(&client->dev, "Error in setting power-down mode\n");
+		dev_err(dev, "Error in setting power-down mode\n");
 		return ret;
 	}
 
Signature · 1 line
2.47.3
#71 lore raw

PATCH v5 18/18 iio: magnetometer: ak8975: make use of the macros from bits.h

In reply to Joshua Crofts

From: Andy Shevchenko <[email protected]>

Make use of BIT() and GENMASK() where it makes sense.

---
 drivers/iio/magnetometer/ak8975.c | 23 ++++++++---------------
 1 file changed, 8 insertions(+), 15 deletions(-)
drivers/iio/magnetometer/ak8975.c +8 -15
index 1c96df3e1ad95d268470609c61be8ee4a709675d..ea024390b36d4a13185ec605de9d808655cab261 100644
@@ -45,8 +45,7 @@
 #define AK8975_REG_INFO			0x01
 
 #define AK8975_REG_ST1			0x02
-#define AK8975_REG_ST1_DRDY_SHIFT	0
-#define AK8975_REG_ST1_DRDY_MASK	(1 << AK8975_REG_ST1_DRDY_SHIFT)
+#define AK8975_REG_ST1_DRDY_MASK	BIT(0)
 
 #define AK8975_REG_HXL			0x03
 #define AK8975_REG_HXH			0x04
@@ -55,15 +54,12 @@
 #define AK8975_REG_HZL			0x07
 #define AK8975_REG_HZH			0x08
 #define AK8975_REG_ST2			0x09
-#define AK8975_REG_ST2_DERR_SHIFT	2
-#define AK8975_REG_ST2_DERR_MASK	(1 << AK8975_REG_ST2_DERR_SHIFT)
+#define AK8975_REG_ST2_DERR_MASK	BIT(2)
 
-#define AK8975_REG_ST2_HOFL_SHIFT	3
-#define AK8975_REG_ST2_HOFL_MASK	(1 << AK8975_REG_ST2_HOFL_SHIFT)
+#define AK8975_REG_ST2_HOFL_MASK	BIT(3)
 
 #define AK8975_REG_CNTL			0x0A
-#define AK8975_REG_CNTL_MODE_SHIFT	0
-#define AK8975_REG_CNTL_MODE_MASK	(0xF << AK8975_REG_CNTL_MODE_SHIFT)
+#define AK8975_REG_CNTL_MODE_MASK	GENMASK(3, 0)
 #define AK8975_REG_CNTL_MODE_POWER_DOWN	0x00
 #define AK8975_REG_CNTL_MODE_ONCE	0x01
 #define AK8975_REG_CNTL_MODE_SELF_TEST	0x08
@@ -95,8 +91,7 @@
 
 #define AK09912_REG_ST1			0x10
 
-#define AK09912_REG_ST1_DRDY_SHIFT	0
-#define AK09912_REG_ST1_DRDY_MASK	(1 << AK09912_REG_ST1_DRDY_SHIFT)
+#define AK09912_REG_ST1_DRDY_MASK	BIT(0)
 
 #define AK09912_REG_HXL			0x11
 #define AK09912_REG_HXH			0x12
@@ -107,8 +102,7 @@
 #define AK09912_REG_TMPS		0x17
 
 #define AK09912_REG_ST2			0x18
-#define AK09912_REG_ST2_HOFL_SHIFT	3
-#define AK09912_REG_ST2_HOFL_MASK	(1 << AK09912_REG_ST2_HOFL_SHIFT)
+#define AK09912_REG_ST2_HOFL_MASK	BIT(3)
 
 #define AK09912_REG_CNTL1		0x30
 
@@ -117,8 +111,7 @@
 #define AK09912_REG_CNTL_MODE_ONCE	0x01
 #define AK09912_REG_CNTL_MODE_SELF_TEST	0x10
 #define AK09912_REG_CNTL_MODE_FUSE_ROM	0x1F
-#define AK09912_REG_CNTL2_MODE_SHIFT	0
-#define AK09912_REG_CNTL2_MODE_MASK	(0x1F << AK09912_REG_CNTL2_MODE_SHIFT)
+#define AK09912_REG_CNTL2_MODE_MASK	GENMASK(4, 0)
 
 #define AK09912_REG_CNTL3		0x32
 
@@ -840,7 +833,7 @@ static const struct iio_chan_spec ak8975_channels[] = {
 	IIO_CHAN_SOFT_TIMESTAMP(3),
 };
 
-static const unsigned long ak8975_scan_masks[] = { 0x7, 0 };
+static const unsigned long ak8975_scan_masks[] = { GENMASK(2, 0), 0 };
 
 static const struct iio_info ak8975_info = {
 	.read_raw = &ak8975_read_raw,
Signature · 1 line
2.47.3
maintainer
#72 lore raw

Re: [PATCH v5 18/18] iio: magnetometer: ak8975: make use of the macros from bits.h

Quoted 7 lines · On Tue, 05 May 2026 13:46:14 +0200 Joshua Crofts via B4 Relay…

From: Andy Shevchenko <[email protected]>

Make use of BIT() and GENMASK() where it makes sense.

One small thing in here. Feel free to just leave that for a separate patch and for now leave it as 0x7

drivers/iio/magnetometer/ak8975.c static const struct iio_chan_spec ak8975_channels[] = { · line 836

Quoted 6 lines
drivers/iio/magnetometer/ak8975.c +1 -1
@@ -840,7 +833,7 @@ static const struct iio_chan_spec ak8975_channels[] = {
 	IIO_CHAN_SOFT_TIMESTAMP(3),
 };
 
-static const unsigned long ak8975_scan_masks[] = { 0x7, 0 };
+static const unsigned long ak8975_scan_masks[] = { GENMASK(2, 0), 0 };

Hmm. This is a fun one... It's actually 3 separate bits so please represent it as BIT(0) | BIT(1) | BIT(2)

or maybe add an enum for the channel scan indexes? BIT(SCAN_X) | BIT(SCAN_Y) | BIT(SCAN_Z) (and there will be an unused here SCAN_TIMESTAMP

Quoted 4 lines
drivers/iio/magnetometer/ak8975.c +0 -0
 static const struct iio_info ak8975_info = {
 	.read_raw = &ak8975_read_raw,
 
#73 lore raw

Re: [PATCH v5 18/18] iio: magnetometer: ak8975: make use of the macros from bits.h

Quoted 2 lines · Jonathan Cameron

On Tue, 05 May 2026 13:46:14 +0200
Joshua Crofts via B4 Relay <[email protected]> wrote:

...

Quoted 4 lines
Quoted 1 line

Make use of BIT() and GENMASK() where it makes sense.

One small thing in here. Feel free to just leave that for a separate patch and for now leave it as 0x7

...

Quoted 4 lines

drivers/iio/magnetometer/ak8975.c static const struct iio_chan_spec ak8975_channels[] = { · line 836

Quoted 2 lines
drivers/iio/magnetometer/ak8975.c +1 -1
-static const unsigned long ak8975_scan_masks[] = { 0x7, 0 };
+static const unsigned long ak8975_scan_masks[] = { GENMASK(2, 0), 0 };

Hmm. This is a fun one... It's actually 3 separate bits so please represent it as BIT(0) | BIT(1) | BIT(2)

Ah, TIL.

Quoted 2 lines

or maybe add an enum for the channel scan indexes? BIT(SCAN_X) | BIT(SCAN_Y) | BIT(SCAN_Z)

This looks good to me.

Quoted 1 line

(and there will be an unused here SCAN_TIMESTAMP

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#74 lore raw

Re: [PATCH v5 18/18] iio: magnetometer: ak8975: make use of the macros from bits.h

Quoted 19 lines · Andy Shevchenko
Quoted 2 lines · Jonathan Cameron

On Tue, 05 May 2026 13:46:14 +0200
Joshua Crofts via B4 Relay <[email protected]> wrote:

...

Quoted 4 lines
Quoted 1 line

Make use of BIT() and GENMASK() where it makes sense.

One small thing in here. Feel free to just leave that for a separate patch and for now leave it as 0x7

...

Quoted 4 lines

drivers/iio/magnetometer/ak8975.c static const struct iio_chan_spec ak8975_channels[] = { · line 836

Quoted 2 lines
drivers/iio/magnetometer/ak8975.c +1 -1
-static const unsigned long ak8975_scan_masks[] = { 0x7, 0 };
+static const unsigned long ak8975_scan_masks[] = { GENMASK(2, 0), 0 };

Hmm. This is a fun one... It's actually 3 separate bits so please represent it as BIT(0) | BIT(1) | BIT(2)

Quoted 6 lines

Ah, TIL.

Quoted 2 lines

or maybe add an enum for the channel scan indexes? BIT(SCAN_X) | BIT(SCAN_Y) | BIT(SCAN_Z)

This looks good to me.

Sure, simple enough.

Signature · 3 lines
Kind regards

CJD