[PATCH] iio: light: gp2ap020a00f: Correct return type to int

[PATCH] iio: light: gp2ap020a00f: Correct return type to int

[Ethan Tidmore]

The function gp2ap020a00f_get_thresh_reg() can return type -EINVAL in
its error path. Yet, the function has return type of u8. Added error
checking for gp2ap020a00f_get_thresh_reg() return's value.

Fixes: 5d6a25bad035 ("iio:gp2ap020a00f: Switch to new event config interface")
Signed-off-by: Ethan Tidmore <ethantidmore06@gmail.com>
---
 drivers/iio/light/gp2ap020a00f.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/drivers/iio/light/gp2ap020a00f.c b/drivers/iio/light/gp2ap020a00f.c
index c7df4b258e2c..492d1a029f77 100644
--- a/drivers/iio/light/gp2ap020a00f.c
+++ b/drivers/iio/light/gp2ap020a00f.c
@@ -992,7 +992,7 @@ static irqreturn_t gp2ap020a00f_trigger_handler(int irq, void *data)
 	return IRQ_HANDLED;
 }
 
-static u8 gp2ap020a00f_get_thresh_reg(const struct iio_chan_spec *chan,
+static int gp2ap020a00f_get_thresh_reg(const struct iio_chan_spec *chan,
 					     enum iio_event_direction event_dir)
 {
 	switch (chan->type) {
@@ -1023,12 +1023,18 @@ static int gp2ap020a00f_write_event_val(struct iio_dev *indio_dev,
 	struct gp2ap020a00f_data *data = iio_priv(indio_dev);
 	bool event_en = false;
 	u8 thresh_val_id;
-	u8 thresh_reg_l;
+	int thresh_reg_l;
 	int err = 0;
 
 	mutex_lock(&data->lock);
 
 	thresh_reg_l = gp2ap020a00f_get_thresh_reg(chan, dir);
+
+	if (thresh_reg_l < 0) {
+		err = thresh_reg_l;
+		goto error_unlock;
+	}
+
 	thresh_val_id = GP2AP020A00F_THRESH_VAL_ID(thresh_reg_l);
 
 	if (thresh_val_id > GP2AP020A00F_THRESH_PH) {
@@ -1080,13 +1086,18 @@ static int gp2ap020a00f_read_event_val(struct iio_dev *indio_dev,
 				       int *val, int *val2)
 {
 	struct gp2ap020a00f_data *data = iio_priv(indio_dev);
-	u8 thresh_reg_l;
+	int thresh_reg_l;
 	int err = IIO_VAL_INT;
 
 	mutex_lock(&data->lock);
 
 	thresh_reg_l = gp2ap020a00f_get_thresh_reg(chan, dir);
 
+	if (thresh_reg_l < 0) {
+		err = thresh_reg_l;
+		goto error_unlock;
+	}
+
 	if (thresh_reg_l > GP2AP020A00F_PH_L_REG) {
 		err = -EINVAL;
 		goto error_unlock;
-- 
2.53.0

Re: [PATCH] iio: light: gp2ap020a00f: Correct return type to int

[Andy Shevchenko]

On Tue, Feb 10, 2026 at 06:48:19PM -0600, Ethan Tidmore wrote:
> The function gp2ap020a00f_get_thresh_reg() can return type -EINVAL in
> its error path. Yet, the function has return type of u8. Added error
> checking for gp2ap020a00f_get_thresh_reg() return's value.

...

> -static u8 gp2ap020a00f_get_thresh_reg(const struct iio_chan_spec *chan,
> +static int gp2ap020a00f_get_thresh_reg(const struct iio_chan_spec *chan,
>  					     enum iio_event_direction event_dir)

While at it, correct the indentation of the second line.

...

>  	mutex_lock(&data->lock);
>  
>  	thresh_reg_l = gp2ap020a00f_get_thresh_reg(chan, dir);

> +

Drop this blank line.

> +	if (thresh_reg_l < 0) {
> +		err = thresh_reg_l;
> +		goto error_unlock;
> +	}

>  	thresh_reg_l = gp2ap020a00f_get_thresh_reg(chan, dir);

>  

Ditto, no blank line between an assignment and a conditional that checks
the value.

> +	if (thresh_reg_l < 0) {
> +		err = thresh_reg_l;
> +		goto error_unlock;
> +	}

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH] iio: light: gp2ap020a00f: Correct return type to int

[Jonathan Cameron]

On Tue, 10 Feb 2026 18:48:19 -0600
Ethan Tidmore <ethantidmore06@gmail.com> wrote:

> The function gp2ap020a00f_get_thresh_reg() can return type -EINVAL in
> its error path. Yet, the function has return type of u8. Added error
> checking for gp2ap020a00f_get_thresh_reg() return's value.
Hi Ethan,

Did you find a path via which it can actually return an error? char
and dir are tightly constrained inputs.

I think the error return is probably just defensive coding against future bugs
and compiler warning supression.

If so it's still a good thing to clean up but Fixes tag and backports
probably not appropriate.

And best of all, once it's not a fix we can use some guard()
magic to make these functions and your new checks cleaner.
guard() usage would be a precursor patch with this one on top to minimize
churn.

Thanks,

Jonathan



> 
> Fixes: 5d6a25bad035 ("iio:gp2ap020a00f: Switch to new event config interface")
> Signed-off-by: Ethan Tidmore <ethantidmore06@gmail.com>
> ---
>  drivers/iio/light/gp2ap020a00f.c | 17 ++++++++++++++---
>  1 file changed, 14 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/iio/light/gp2ap020a00f.c b/drivers/iio/light/gp2ap020a00f.c
> index c7df4b258e2c..492d1a029f77 100644
> --- a/drivers/iio/light/gp2ap020a00f.c
> +++ b/drivers/iio/light/gp2ap020a00f.c
> @@ -992,7 +992,7 @@ static irqreturn_t gp2ap020a00f_trigger_handler(int irq, void *data)
>  	return IRQ_HANDLED;
>  }
>  
> -static u8 gp2ap020a00f_get_thresh_reg(const struct iio_chan_spec *chan,
> +static int gp2ap020a00f_get_thresh_reg(const struct iio_chan_spec *chan,
>  					     enum iio_event_direction event_dir)
>  {
>  	switch (chan->type) {
> @@ -1023,12 +1023,18 @@ static int gp2ap020a00f_write_event_val(struct iio_dev *indio_dev,
>  	struct gp2ap020a00f_data *data = iio_priv(indio_dev);
>  	bool event_en = false;
>  	u8 thresh_val_id;
> -	u8 thresh_reg_l;
> +	int thresh_reg_l;
>  	int err = 0;
>  
>  	mutex_lock(&data->lock);
>  
>  	thresh_reg_l = gp2ap020a00f_get_thresh_reg(chan, dir);
> +
> +	if (thresh_reg_l < 0) {
> +		err = thresh_reg_l;
> +		goto error_unlock;
> +	}
> +
>  	thresh_val_id = GP2AP020A00F_THRESH_VAL_ID(thresh_reg_l);
>  
>  	if (thresh_val_id > GP2AP020A00F_THRESH_PH) {
> @@ -1080,13 +1086,18 @@ static int gp2ap020a00f_read_event_val(struct iio_dev *indio_dev,
>  				       int *val, int *val2)
>  {
>  	struct gp2ap020a00f_data *data = iio_priv(indio_dev);
> -	u8 thresh_reg_l;
> +	int thresh_reg_l;
>  	int err = IIO_VAL_INT;
>  
>  	mutex_lock(&data->lock);
>  
>  	thresh_reg_l = gp2ap020a00f_get_thresh_reg(chan, dir);
>  
> +	if (thresh_reg_l < 0) {
> +		err = thresh_reg_l;
> +		goto error_unlock;
> +	}
> +
>  	if (thresh_reg_l > GP2AP020A00F_PH_L_REG) {
>  		err = -EINVAL;
>  		goto error_unlock;

Re: [PATCH] iio: light: gp2ap020a00f: Correct return type to int

[Ethan Tidmore]

On Sat Feb 14, 2026 at 1:08 PM CST, Jonathan Cameron wrote:
> On Tue, 10 Feb 2026 18:48:19 -0600
> Ethan Tidmore <ethantidmore06@gmail.com> wrote:

...

> Did you find a path via which it can actually return an error? char
> and dir are tightly constrained inputs.
>
> I think the error return is probably just defensive coding against future bugs
> and compiler warning supression.

Yes you're right. I was just correcting a signness warning from smatch
and just added the extra checks.

>
> If so it's still a good thing to clean up but Fixes tag and backports
> probably not appropriate.
>
> And best of all, once it's not a fix we can use some guard()
> magic to make these functions and your new checks cleaner.
> guard() usage would be a precursor patch with this one on top to minimize
> churn.

I'll send a series adding these checks and using guard() and will drop
the Fixes: tag.

Thanks, 

ET