[PATCH 0/3] gpiolib: fix bugs in retval sanitization

[PATCH 0/3] gpiolib: fix bugs in retval sanitization

[Bartosz Golaszewski]

Here's a set of fixes to issues spotted in next after queuing the series
adding return value sanitization to GPIO core.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
---
Bartosz Golaszewski (3):
      gpiolib: don't use gpiochip_get_direction() when registering a chip
      gpiolib: use a more explicit retval logic in gpiochip_get_direction()
      gpiolib: don't double-check the gc->get callback's existence

 drivers/gpio/gpiolib.c | 29 ++++++++---------------------
 1 file changed, 8 insertions(+), 21 deletions(-)
---
base-commit: 0226d0ce98a477937ed295fb7df4cc30b46fc304
change-id: 20250225-retval-fixes-a1a09a1e3a08

Best regards,
-- 
Bartosz Golaszewski <bartosz.golaszewski@linaro.org>

[PATCH 1/3] gpiolib: don't use gpiochip_get_direction() when registering a chip

[Bartosz Golaszewski]

From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>

During chip registration we should neither check the return value of
gc->get_direction() nor hold the SRCU lock when calling it. The former
is because pin controllers may have pins set to alternate functions and
return errors from their get_direction() callbacks. That's alright - we
should default to the safe INPUT state and not bail-out. The latter is
not needed because we haven't registered the chip yet so there's nothing
to protect against dynamic removal. In fact: we currently hit a lockdep
splat. Revert to calling the gc->get_direction() callback directly not
not checking its value.

Fixes: 9d846b1aebbe ("gpiolib: check the return value of gpio_chip::get_direction()")
Fixes: e623c4303ed1 ("gpiolib: sanitize the return value of gpio_chip::get_direction()")
Reported-by: Marek Szyprowski <m.szyprowski@samsung.com>
Closes: https://lore.kernel.org/all/81f890fc-6688-42f0-9756-567efc8bb97a@samsung.com/
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
---
 drivers/gpio/gpiolib.c | 20 ++++----------------
 1 file changed, 4 insertions(+), 16 deletions(-)

diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c
index e8678a6c82ea..31d400b10167 100644
--- a/drivers/gpio/gpiolib.c
+++ b/drivers/gpio/gpiolib.c
@@ -1082,24 +1082,12 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,
 
 		desc->gdev = gdev;
 
-		if (gc->get_direction && gpiochip_line_is_valid(gc, desc_index)) {
-			ret = gpiochip_get_direction(gc, desc_index);
-			if (ret < 0)
-				/*
-				 * FIXME: Bail-out here once all GPIO drivers
-				 * are updated to not return errors in
-				 * situations that can be considered normal
-				 * operation.
-				 */
-				dev_warn(&gdev->dev,
-					 "%s: get_direction failed: %d\n",
-					 __func__, ret);
-
-			assign_bit(FLAG_IS_OUT, &desc->flags, !ret);
-		} else {
+		if (gc->get_direction && gpiochip_line_is_valid(gc, desc_index))
+			assign_bit(FLAG_IS_OUT, &desc->flags,
+				   !gc->get_direction(gc, desc_index));
+		else
 			assign_bit(FLAG_IS_OUT,
 				   &desc->flags, !gc->direction_input);
-		}
 	}
 
 	ret = of_gpiochip_add(gc);

-- 
2.45.2

Re: [PATCH 1/3] gpiolib: don't use gpiochip_get_direction() when registering a chip

[Marek Szyprowski]

On 25.02.2025 12:56, Bartosz Golaszewski wrote:
> From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
>
> During chip registration we should neither check the return value of
> gc->get_direction() nor hold the SRCU lock when calling it. The former
> is because pin controllers may have pins set to alternate functions and
> return errors from their get_direction() callbacks. That's alright - we
> should default to the safe INPUT state and not bail-out. The latter is
> not needed because we haven't registered the chip yet so there's nothing
> to protect against dynamic removal. In fact: we currently hit a lockdep
> splat. Revert to calling the gc->get_direction() callback directly not
> not checking its value.
>
> Fixes: 9d846b1aebbe ("gpiolib: check the return value of gpio_chip::get_direction()")
> Fixes: e623c4303ed1 ("gpiolib: sanitize the return value of gpio_chip::get_direction()")
> Reported-by: Marek Szyprowski <m.szyprowski@samsung.com>
> Closes: https://lore.kernel.org/all/81f890fc-6688-42f0-9756-567efc8bb97a@samsung.com/
> Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
> ---
>   drivers/gpio/gpiolib.c | 20 ++++----------------
>   1 file changed, 4 insertions(+), 16 deletions(-)
>
> diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c
> index e8678a6c82ea..31d400b10167 100644
> --- a/drivers/gpio/gpiolib.c
> +++ b/drivers/gpio/gpiolib.c
> @@ -1082,24 +1082,12 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,
>   
>   		desc->gdev = gdev;
>   
> -		if (gc->get_direction && gpiochip_line_is_valid(gc, desc_index)) {
> -			ret = gpiochip_get_direction(gc, desc_index);
> -			if (ret < 0)
> -				/*
> -				 * FIXME: Bail-out here once all GPIO drivers
> -				 * are updated to not return errors in
> -				 * situations that can be considered normal
> -				 * operation.
> -				 */
> -				dev_warn(&gdev->dev,
> -					 "%s: get_direction failed: %d\n",
> -					 __func__, ret);
> -
> -			assign_bit(FLAG_IS_OUT, &desc->flags, !ret);
> -		} else {
> +		if (gc->get_direction && gpiochip_line_is_valid(gc, desc_index))
> +			assign_bit(FLAG_IS_OUT, &desc->flags,
> +				   !gc->get_direction(gc, desc_index));
> +		else
>   			assign_bit(FLAG_IS_OUT,
>   				   &desc->flags, !gc->direction_input);
> -		}
>   	}
>   
>   	ret = of_gpiochip_add(gc);
>
Best regards
-- 
Marek Szyprowski, PhD
Samsung R&D Institute Poland

Re: [PATCH 1/3] gpiolib: don't use gpiochip_get_direction() when registering a chip

[Andy Shevchenko]

On Tue, Feb 25, 2025 at 12:56:23PM +0100, Bartosz Golaszewski wrote:
> From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
> 
> During chip registration we should neither check the return value of
> gc->get_direction() nor hold the SRCU lock when calling it. The former
> is because pin controllers may have pins set to alternate functions and
> return errors from their get_direction() callbacks. That's alright - we
> should default to the safe INPUT state and not bail-out. The latter is
> not needed because we haven't registered the chip yet so there's nothing
> to protect against dynamic removal. In fact: we currently hit a lockdep
> splat. Revert to calling the gc->get_direction() callback directly not
> not checking its value.

...

I think the below code deserves a commit (as a summary of the above commit
message).

> +		if (gc->get_direction && gpiochip_line_is_valid(gc, desc_index))
> +			assign_bit(FLAG_IS_OUT, &desc->flags,
> +				   !gc->get_direction(gc, desc_index));
> +		else
>  			assign_bit(FLAG_IS_OUT,
>  				   &desc->flags, !gc->direction_input);

Otherwise LGTM,
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH 1/3] gpiolib: don't use gpiochip_get_direction() when registering a chip

[Bartosz Golaszewski]

On Tue, Feb 25, 2025 at 2:22 PM Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
>
> On Tue, Feb 25, 2025 at 12:56:23PM +0100, Bartosz Golaszewski wrote:
> > From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
> >
> > During chip registration we should neither check the return value of
> > gc->get_direction() nor hold the SRCU lock when calling it. The former
> > is because pin controllers may have pins set to alternate functions and
> > return errors from their get_direction() callbacks. That's alright - we
> > should default to the safe INPUT state and not bail-out. The latter is
> > not needed because we haven't registered the chip yet so there's nothing
> > to protect against dynamic removal. In fact: we currently hit a lockdep
> > splat. Revert to calling the gc->get_direction() callback directly not
> > not checking its value.
>
> ...
>
> I think the below code deserves a commit (as a summary of the above commit
> message).
>

Can you rephrase? I'm not getting this one.

Bart

> > +             if (gc->get_direction && gpiochip_line_is_valid(gc, desc_index))
> > +                     assign_bit(FLAG_IS_OUT, &desc->flags,
> > +                                !gc->get_direction(gc, desc_index));
> > +             else
> >                       assign_bit(FLAG_IS_OUT,
> >                                  &desc->flags, !gc->direction_input);
>
> Otherwise LGTM,
> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
>
> --
> With Best Regards,
> Andy Shevchenko
>
>

Re: [PATCH 1/3] gpiolib: don't use gpiochip_get_direction() when registering a chip

[Andy Shevchenko]

On Tue, Feb 25, 2025 at 03:43:29PM +0100, Bartosz Golaszewski wrote:
> On Tue, Feb 25, 2025 at 2:22 PM Andy Shevchenko
> <andriy.shevchenko@linux.intel.com> wrote:
> > On Tue, Feb 25, 2025 at 12:56:23PM +0100, Bartosz Golaszewski wrote:

> > > During chip registration we should neither check the return value of
> > > gc->get_direction() nor hold the SRCU lock when calling it. The former
> > > is because pin controllers may have pins set to alternate functions and
> > > return errors from their get_direction() callbacks. That's alright - we
> > > should default to the safe INPUT state and not bail-out. The latter is
> > > not needed because we haven't registered the chip yet so there's nothing
> > > to protect against dynamic removal. In fact: we currently hit a lockdep
> > > splat. Revert to calling the gc->get_direction() callback directly not
> > > not checking its value.

...

> > I think the below code deserves a commit (as a summary of the above commit
> > message).
> 
> Can you rephrase? I'm not getting this one.

Ah, s/commit/comment/

> > > +             if (gc->get_direction && gpiochip_line_is_valid(gc, desc_index))
> > > +                     assign_bit(FLAG_IS_OUT, &desc->flags,
> > > +                                !gc->get_direction(gc, desc_index));
> > > +             else
> > >                       assign_bit(FLAG_IS_OUT,
> > >                                  &desc->flags, !gc->direction_input);
> >
> > Otherwise LGTM,
> > Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>

-- 
With Best Regards,
Andy Shevchenko

[PATCH 2/3] gpiolib: use a more explicit retval logic in gpiochip_get_direction()

[Bartosz Golaszewski]

From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>

We have existing macros for direction settings so we don't need to rely
on the magic value of 1 in the retval check. Use readable logic that
explicitly says we expect INPUT, OUTPUT or a negative errno and nothing
else in gpiochip_get_direction().

Fixes: e623c4303ed1 ("gpiolib: sanitize the return value of gpio_chip::get_direction()")
Suggested-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Closes: https://lore.kernel.org/all/Z7yfTggRrk3K6srs@black.fi.intel.com/
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
---
 drivers/gpio/gpiolib.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c
index 31d400b10167..d076b2ec633f 100644
--- a/drivers/gpio/gpiolib.c
+++ b/drivers/gpio/gpiolib.c
@@ -352,7 +352,8 @@ static int gpiochip_get_direction(struct gpio_chip *gc, unsigned int offset)
 		return -EOPNOTSUPP;
 
 	ret = gc->get_direction(gc, offset);
-	if (ret > 1)
+	if (!(ret == GPIO_LINE_DIRECTION_OUT ||
+	      ret == GPIO_LINE_DIRECTION_IN || ret < 0))
 		ret = -EBADE;
 
 	return ret;

-- 
2.45.2

Re: [PATCH 2/3] gpiolib: use a more explicit retval logic in gpiochip_get_direction()

[Andy Shevchenko]

On Tue, Feb 25, 2025 at 12:56:24PM +0100, Bartosz Golaszewski wrote:
> From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
> 
> We have existing macros for direction settings so we don't need to rely
> on the magic value of 1 in the retval check. Use readable logic that
> explicitly says we expect INPUT, OUTPUT or a negative errno and nothing
> else in gpiochip_get_direction().

...

>  	ret = gc->get_direction(gc, offset);
> -	if (ret > 1)
> +	if (!(ret == GPIO_LINE_DIRECTION_OUT ||
> +	      ret == GPIO_LINE_DIRECTION_IN || ret < 0))
>  		ret = -EBADE;

Wouldn't be better to write it as

	if (ret < 0)
		return ret;

	if (ret != GPIO_LINE_DIRECTION_OUT && ret != GPIO_LINE_DIRECTION_IN)
		ret = -EBADE;

	return ret;


Otherwise LGTM,
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
after addressing the above.

-- 
With Best Regards,
Andy Shevchenko

[PATCH 3/3] gpiolib: don't double-check the gc->get callback's existence

[Bartosz Golaszewski]

From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>

gpiochip_get() is called only in two places: in gpio_chip_get_value()
and in gpiochip_get_multiple() where the existence of the gc->get()
callback is already checked. It makes sense to unduplicate the check by
moving it one level up the stack.

Fixes: 86ef402d805d ("gpiolib: sanitize the return value of gpio_chip::get()")
Suggested-by: Andy Shevchenko <andriy.shevchenko@intel.com>
Closes: https://lore.kernel.org/all/Z7yekJ8uRh8dphKn@black.fi.intel.com/
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
---
 drivers/gpio/gpiolib.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c
index d076b2ec633f..b8f10192f27e 100644
--- a/drivers/gpio/gpiolib.c
+++ b/drivers/gpio/gpiolib.c
@@ -3158,9 +3158,7 @@ static int gpiochip_get(struct gpio_chip *gc, unsigned int offset)
 
 	lockdep_assert_held(&gc->gpiodev->srcu);
 
-	if (!gc->get)
-		return -EIO;
-
+	/* Make sure this is called after checking for gc->get(). */
 	ret = gc->get(gc, offset);
 	if (ret > 1)
 		ret = -EBADE;
@@ -3170,7 +3168,7 @@ static int gpiochip_get(struct gpio_chip *gc, unsigned int offset)
 
 static int gpio_chip_get_value(struct gpio_chip *gc, const struct gpio_desc *desc)
 {
-	return gpiochip_get(gc, gpio_chip_hwgpio(desc));
+	return gc->get ? gpiochip_get(gc, gpio_chip_hwgpio(desc)) : -EIO;
 }
 
 /* I/O calls are only valid after configuration completed; the relevant

-- 
2.45.2

Re: [PATCH 3/3] gpiolib: don't double-check the gc->get callback's existence

[Andy Shevchenko]

On Tue, Feb 25, 2025 at 12:56:25PM +0100, Bartosz Golaszewski wrote:
> From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
> 
> gpiochip_get() is called only in two places: in gpio_chip_get_value()
> and in gpiochip_get_multiple() where the existence of the gc->get()
> callback is already checked. It makes sense to unduplicate the check by
> moving it one level up the stack.

Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>

-- 
With Best Regards,
Andy Shevchenko