iio: st_lsm6dsx: Replace scnprintf with sysfs_emit

PATCH Superseded Akshay Bansod 5 participants 10 messages Jul 2, 2025 – Jul 3, 2025

Subsystems (2)

IIO ST LSM6DSx IMU IIO driver

lore

Share & tools

lore

Akshay Bansod

author

2025-07-02 13:59 UTC #0 lore raw

PATCH 1/1 iio: st_lsm6dsx: Replace scnprintf with sysfs_emit

Update the sysfs interface for sampling frequency and scale attributes. Replace scnprintf() with sysfs_emit_at() which is PAGE_SIZE-aware and recommended for use in sysfs.

Signed-off-by Akshay Bansod <[email protected]>

---

Testing:
- Built the driver (st_lsm6dsx_i2c) as a module.
- Tested using i2c-stub to mock the device.
- Verified that reading sysfs attributes like sampling_frequency_available works correctly and shows no change in functionality.

 drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +2 -2

index c65ad4982..1cef10919 100644
@@ -2035,7 +2035,7 @@ st_lsm6dsx_sysfs_sampling_frequency_avail(struct device *dev,
 
 	odr_table = &sensor->hw->settings->odr_table[sensor->id];
 	for (i = 0; i < odr_table->odr_len; i++)
-		len += scnprintf(buf + len, PAGE_SIZE - len, "%d.%03d ",
+		len += sysfs_emit_at(buf, len, "%d.%03d ",
 				 odr_table->odr_avl[i].milli_hz / 1000,
 				 odr_table->odr_avl[i].milli_hz % 1000);
 	buf[len - 1] = '\n';

@@ -2054,7 +2054,7 @@ static ssize_t st_lsm6dsx_sysfs_scale_avail(struct device *dev,
 
 	fs_table = &hw->settings->fs_table[sensor->id];
 	for (i = 0; i < fs_table->fs_len; i++)
-		len += scnprintf(buf + len, PAGE_SIZE - len, "0.%09u ",
+		len += sysfs_emit_at(buf, len, "0.%09u ",
 				 fs_table->fs_avl[i].gain);
 	buf[len - 1] = '\n';

Signature · 1 line

2.49.0

David Lechner

reviewer

2025-07-02 14:16 UTC #1 lore raw

Quoted 4 lines · Akshay Bansod

Update the sysfs interface for sampling frequency and scale attributes. Replace scnprintf() with sysfs_emit_at() which is PAGE_SIZE-aware and recommended for use in sysfs.

Quoted 8 lines

Signed-off-by Akshay Bansod <[email protected]>
---
Testing:
- Built the driver (st_lsm6dsx_i2c) as a module.
- Tested using i2c-stub to mock the device.
- Verified that reading sysfs attributes like sampling_frequency_available works correctly and shows no change in functionality.

Nice to see it was actually tested. :-)

Quoted 9 lines

 drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +0 -0

index c65ad4982..1cef10919 100644

drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c st_lsm6dsx_sysfs_sampling_frequency_avail(struct device *dev, · line 2040

Quoted 7 lines

drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +1 -1

@@ -2035,7 +2035,7 @@ st_lsm6dsx_sysfs_sampling_frequency_avail(struct device *dev,
 
 	odr_table = &sensor->hw->settings->odr_table[sensor->id];
 	for (i = 0; i < odr_table->odr_len; i++)
-		len += scnprintf(buf + len, PAGE_SIZE - len, "%d.%03d ",
+		len += sysfs_emit_at(buf, len, "%d.%03d ",
 				 odr_table->odr_avl[i].milli_hz / 1000,
 				 odr_table->odr_avl[i].milli_hz % 1000);

Let's keep checkpatch happy and change the indent of the wrapped lines to line up with ( since the ( moved.

drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c static ssize_t st_lsm6dsx_sysfs_scale_avail(struct device *dev, · line 2058

Quoted 8 lines

drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +1 -1

 	buf[len - 1] = '\n';

@@ -2054,7 +2054,7 @@ static ssize_t st_lsm6dsx_sysfs_scale_avail(struct device *dev,
 
 	fs_table = &hw->settings->fs_table[sensor->id];
 	for (i = 0; i < fs_table->fs_len; i++)
-		len += scnprintf(buf + len, PAGE_SIZE - len, "0.%09u ",
+		len += sysfs_emit_at(buf, len, "0.%09u ",
 				 fs_table->fs_avl[i].gain);

ditto

Quoted 2 lines

drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +0 -0
 	buf[len - 1] = '\n';
 

Andy Shevchenko

2025-07-02 14:55 UTC #2 lore raw

Quoted 4 lines · David Lechner

Quoted 3 lines · Akshay Bansod
Update the sysfs interface for sampling frequency and scale attributes. Replace scnprintf() with sysfs_emit_at() which is PAGE_SIZE-aware and recommended for use in sysfs.

...

Quoted 6 lines

drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c st_lsm6dsx_sysfs_sampling_frequency_avail(struct device *dev, · line 2040
Quoted 3 lines
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +1 -0
+		len += sysfs_emit_at(buf, len, "%d.%03d ",
 				 odr_table->odr_avl[i].milli_hz / 1000,
 				 odr_table->odr_avl[i].milli_hz % 1000);
Let's keep checkpatch happy and change the indent of the wrapped lines to line up with ( since the ( moved.

While I see the point, wouldn't be better to have 1000 replaced with MILLI at the same time?

Signature · 2 lines

With Best Regards,
Andy Shevchenko

David Lechner

reviewer

2025-07-02 15:04 UTC #3 lore raw

Quoted 15 lines · Andy Shevchenko

Quoted 4 lines · David Lechner
Quoted 3 lines · Akshay Bansod
Update the sysfs interface for sampling frequency and scale attributes. Replace scnprintf() with sysfs_emit_at() which is PAGE_SIZE-aware and recommended for use in sysfs.
...
Quoted 6 lines
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c st_lsm6dsx_sysfs_sampling_frequency_avail(struct device *dev, · line 2040
Quoted 3 lines
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +1 -0
+		len += sysfs_emit_at(buf, len, "%d.%03d ",
 				 odr_table->odr_avl[i].milli_hz / 1000,
 				 odr_table->odr_avl[i].milli_hz % 1000);
Let's keep checkpatch happy and change the indent of the wrapped lines to line up with ( since the ( moved.

Quoted 3 lines

While I see the point, wouldn't be better to have 1000 replaced with MILLI at the same time?

For anything with 3 zeros, I don't consider MILLI better (or worse). Science shows that the average human can easily see 3 or 4 things without having to count them [1]. So it is only when we start getting more 0s than that is when I think we should be picky about using macros instead.

And in this particular case, we are converting milli to micro so 1000 should be replaced by (MICRO / MILLI) if we are going to do that.

[1]: https://www.scientificamerican.com/article/your-brain-finds-it-easy-to-size-up-four-objects-but-not-five-heres-why/

Andy Shevchenko

2025-07-02 15:22 UTC #4 lore raw

Quoted 3 lines · David Lechner

Quoted 2 lines · Andy Shevchenko
Quoted 1 line · David Lechner
On 7/2/25 8:58 AM, Akshay Bansod wrote:

...

Quoted 10 lines

Quoted 9 lines
Quoted 6 lines
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c st_lsm6dsx_sysfs_sampling_frequency_avail(struct device *dev, · line 2040
Quoted 3 lines
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +1 -0
+		len += sysfs_emit_at(buf, len, "%d.%03d ",
 				 odr_table->odr_avl[i].milli_hz / 1000,
 				 odr_table->odr_avl[i].milli_hz % 1000);
Let's keep checkpatch happy and change the indent of the wrapped lines to line up with ( since the ( moved.
While I see the point, wouldn't be better to have 1000 replaced with MILLI at the same time?

Quoted 8 lines

For anything with 3 zeros, I don't consider MILLI better (or worse). Science shows that the average human can easily see 3 or 4 things without having to count them [1]. So it is only when we start getting more 0s than that is when I think we should be picky about using macros instead.
And in this particular case, we are converting milli to micro so 1000 should be replaced by (MICRO / MILLI) if we are going to do that.

I see. This changes the picture drastically. Let's leave it for another day then.

Quoted 1 line

[1]: https://www.scientificamerican.com/article/your-brain-finds-it-easy-to-size-up-four-objects-but-not-five-heres-why/

Signature · 2 lines

With Best Regards,
Andy Shevchenko

Jonathan Cameron

2025-07-02 15:33 UTC #5 lore raw

In reply to David Lechner

Quoted 20 lines · On Wed, 2 Jul 2025 10:04:23 -0500 David Lechner <[email protected]> wrote:

Quoted 15 lines · Andy Shevchenko
Quoted 4 lines · David Lechner
Quoted 3 lines · Akshay Bansod
Update the sysfs interface for sampling frequency and scale attributes. Replace scnprintf() with sysfs_emit_at() which is PAGE_SIZE-aware and recommended for use in sysfs.
...
Quoted 6 lines
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c st_lsm6dsx_sysfs_sampling_frequency_avail(struct device *dev, · line 2040
Quoted 3 lines
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +1 -0
+		len += sysfs_emit_at(buf, len, "%d.%03d ",
 				 odr_table->odr_avl[i].milli_hz / 1000,
 				 odr_table->odr_avl[i].milli_hz % 1000);  
Let's keep checkpatch happy and change the indent of the wrapped lines to line up with ( since the ( moved.
Quoted 3 lines
While I see the point, wouldn't be better to have 1000 replaced with MILLI at the same time?

Quoted 8 lines

For anything with 3 zeros, I don't consider MILLI better (or worse). Science shows that the average human can easily see 3 or 4 things without having to count them [1]. So it is only when we start getting more 0s than that is when I think we should be picky about using macros instead.
And in this particular case, we are converting milli to micro so 1000 should be replaced by (MICRO / MILLI) if we are going to do that.

No we aren't.

This one is converting from milli_hz to hz + sticking to milli for the decimal part.

Lots of other IIO cases where you would have been right, but I think not here.

Quoted 4 lines

[1]: https://www.scientificamerican.com/article/your-brain-finds-it-easy-to-size-up-four-objects-but-not-five-heres-why/

David Lechner

reviewer

2025-07-02 15:53 UTC #6 lore raw

Quoted 29 lines · Jonathan Cameron

Quoted 26 lines · On Wed, 2 Jul 2025 10:04:23 -0500 David Lechner <[email protected]> wrote:
Quoted 15 lines · Andy Shevchenko
Quoted 4 lines · David Lechner
Quoted 3 lines · Akshay Bansod
Update the sysfs interface for sampling frequency and scale attributes. Replace scnprintf() with sysfs_emit_at() which is PAGE_SIZE-aware and recommended for use in sysfs.
...
Quoted 6 lines
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c st_lsm6dsx_sysfs_sampling_frequency_avail(struct device *dev, · line 2040
Quoted 3 lines
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +1 -0
+		len += sysfs_emit_at(buf, len, "%d.%03d ",
 				 odr_table->odr_avl[i].milli_hz / 1000,
 				 odr_table->odr_avl[i].milli_hz % 1000);  
Let's keep checkpatch happy and change the indent of the wrapped lines to line up with ( since the ( moved.
Quoted 3 lines
While I see the point, wouldn't be better to have 1000 replaced with MILLI at the same time?
For anything with 3 zeros, I don't consider MILLI better (or worse). Science shows that the average human can easily see 3 or 4 things without having to count them [1]. So it is only when we start getting more 0s than that is when I think we should be picky about using macros instead.

Quoted 8 lines

Quoted 2 lines
And in this particular case, we are converting milli to micro so 1000 should be replaced by (MICRO / MILLI) if we are going to do that.
No we aren't.
This one is converting from milli_hz to hz + sticking to milli for the decimal part.
Lots of other IIO cases where you would have been right, but I think not here.

Oops. The %03d instead of %06d should have given it away!

Quoted 5 lines

Quoted 4 lines
[1]: https://www.scientificamerican.com/article/your-brain-finds-it-easy-to-size-up-four-objects-but-not-five-heres-why/

Andy Shevchenko

2025-07-03 09:05 UTC #7 lore raw

Wed, Jul 02, 2025 at 10:53:31AM -0500, David Lechner kirjoitti:

Quoted 6 lines · Jonathan Cameron

Quoted 5 lines
Quoted 3 lines · On Wed, 2 Jul 2025 10:04:23 -0500 David Lechner <[email protected]> wrote:
Quoted 2 lines · Andy Shevchenko
Quoted 1 line · David Lechner
On 7/2/25 8:58 AM, Akshay Bansod wrote:

...

Quoted 18 lines

Quoted 18 lines
Quoted 18 lines
Quoted 9 lines
Quoted 6 lines
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c st_lsm6dsx_sysfs_sampling_frequency_avail(struct device *dev, · line 2040
Quoted 3 lines
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +1 -0
+		len += sysfs_emit_at(buf, len, "%d.%03d ",
 				 odr_table->odr_avl[i].milli_hz / 1000,
 				 odr_table->odr_avl[i].milli_hz % 1000);  
Let's keep checkpatch happy and change the indent of the wrapped lines to line up with ( since the ( moved.
While I see the point, wouldn't be better to have 1000 replaced with MILLI at the same time?
For anything with 3 zeros, I don't consider MILLI better (or worse). Science shows that the average human can easily see 3 or 4 things without having to count them [1]. So it is only when we start getting more 0s than that is when I think we should be picky about using macros instead.
And in this particular case, we are converting milli to micro so 1000 should be replaced by (MICRO / MILLI) if we are going to do that.

Quoted 8 lines

Quoted 6 lines
No we aren't.
This one is converting from milli_hz to hz + sticking to milli for the decimal part.
Lots of other IIO cases where you would have been right, but I think not here.
Oops. The %03d instead of %06d should have given it away!

I'm not sure I got your comment. The '3' vs. '6' will just define the minimum amount of printed digits, it does *not* limit the upper numbers anyhow (it's limited by the 'd', which is (INT_MIN .. INT_MAX).

Quoted 1 line

Quoted 1 line
Quoted 1 line
[1]: https://www.scientificamerican.com/article/your-brain-finds-it-easy-to-size-up-four-objects-but-not-five-heres-why/

Signature · 2 lines

With Best Regards,
Andy Shevchenko

David Lechner

reviewer

2025-07-03 12:31 UTC #8 lore raw

Quoted 33 lines · Andy Shevchenko

Wed, Jul 02, 2025 at 10:53:31AM -0500, David Lechner kirjoitti:
Quoted 6 lines · Jonathan Cameron
Quoted 5 lines
Quoted 3 lines · On Wed, 2 Jul 2025 10:04:23 -0500 David Lechner <[email protected]> wrote:
Quoted 2 lines · Andy Shevchenko
Quoted 1 line · David Lechner
On 7/2/25 8:58 AM, Akshay Bansod wrote:
...
Quoted 23 lines
Quoted 23 lines
Quoted 10 lines
Quoted 9 lines
Quoted 6 lines
Review
Quoted 3 lines
(quoted diff) +1 -0
+		len += sysfs_emit_at(buf, len, "%d.%03d ",
 				 odr_table->odr_avl[i].milli_hz / 1000,
 				 odr_table->odr_avl[i].milli_hz % 1000);  
Let's keep checkpatch happy and change the indent of the wrapped lines to line up with ( since the ( moved.
While I see the point, wouldn't be better to have 1000 replaced with MILLI at the same time?
Quoted 8 lines
For anything with 3 zeros, I don't consider MILLI better (or worse). Science shows that the average human can easily see 3 or 4 things without having to count them [1]. So it is only when we start getting more 0s than that is when I think we should be picky about using macros instead.
And in this particular case, we are converting milli to micro so 1000 should be replaced by (MICRO / MILLI) if we are going to do that.
No we aren't.
This one is converting from milli_hz to hz + sticking to milli for the decimal part.

Quoted 7 lines

Quoted 3 lines
Quoted 1 line
Lots of other IIO cases where you would have been right, but I think not here.
Oops. The %03d instead of %06d should have given it away!
I'm not sure I got your comment. The '3' vs. '6' will just define the minimum amount of printed digits, it does *not* limit the upper numbers anyhow (it's limited by the 'd', which is (INT_MIN .. INT_MAX).

It is after the decimal point in the printed string, so 3 digits after a decimal point is going to be MILLI units. And the % 1000 ensures that we would never get more than 3 digits there.

Quoted 4 lines

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
[1]: https://www.scientificamerican.com/article/your-brain-finds-it-easy-to-size-up-four-objects-but-not-five-heres-why/

akshay bansod

author

2025-07-02 14:58 UTC #9 lore raw

In reply to David Lechner

Quoted 8 lines · On Wednesday, 2 July 2025 8:25 pm +0530 David Lechner wrote:

Quoted 7 lines · Akshay Bansod
Update the sysfs interface for sampling frequency and scale attributes. Replace scnprintf() with sysfs_emit_at() which is PAGE_SIZE-aware and recommended for use in sysfs.
Signed-off-by Akshay Bansod <[email protected]>
---

Quoted 8 lines

Quoted 5 lines
Testing:
- Built the driver (st_lsm6dsx_i2c) as a module.
- Tested using i2c-stub to mock the device.
- Verified that reading sysfs attributes like sampling_frequency_available works correctly and shows no change in functionality.
Nice to see it was actually tested. :-)

:-)

Quoted 10 lines

 drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +0 -0

index c65ad4982..1cef10919 100644
@@ -2035,7 +2035,7 @@ st_lsm6dsx_sysfs_sampling_frequency_avail(struct device *dev,

Quoted 10 lines

drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c st_lsm6dsx_sysfs_sampling_frequency_avail(struct device *dev, · line 2040
Quoted 6 lines
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +1 -1
 	odr_table = &sensor->hw->settings->odr_table[sensor->id];
 	for (i = 0; i < odr_table->odr_len; i++)
-		len += scnprintf(buf + len, PAGE_SIZE - len, "%d.%03d ",
+		len += sysfs_emit_at(buf, len, "%d.%03d ",
 				 odr_table->odr_avl[i].milli_hz / 1000,
 				 odr_table->odr_avl[i].milli_hz % 1000);
Let's keep checkpatch happy and change the indent of the wrapped lines to line up with ( since the ( moved.

noted. I wasn't aware of that.

Quoted 11 lines

drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c static ssize_t st_lsm6dsx_sysfs_scale_avail(struct device *dev, · line 2058
Quoted 8 lines
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +1 -1
 	buf[len - 1] = '\n';
@@ -2054,7 +2054,7 @@ static ssize_t st_lsm6dsx_sysfs_scale_avail(struct device *dev,
 
 	fs_table = &hw->settings->fs_table[sensor->id];
 	for (i = 0; i < fs_table->fs_len; i++)
-		len += scnprintf(buf + len, PAGE_SIZE - len, "0.%09u ",
+		len += sysfs_emit_at(buf, len, "0.%09u ",
 				 fs_table->fs_avl[i].gain);
ditto

noted.

Quoted 4 lines

drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c static ssize_t st_lsm6dsx_sysfs_scale_avail(struct device *dev, · line 2060
Quoted 2 lines
drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c +0 -0
 	buf[len - 1] = '\n';
 

Thanks for the review. I'll revise the patch.

Regards,
Akshay