[PATCH v3] drivers/iio: Remove all strcpy() uses

[PATCH v3] drivers/iio: Remove all strcpy() uses

[Len Baker]

strcpy() performs no bounds checking on the destination buffer. This
could result in linear overflows beyond the end of the buffer, leading
to all kinds of misbehaviors. So, remove all the uses and add
devm_kstrdup() or devm_kasprintf() instead.

This patch is an effort to clean up the proliferation of str*()
functions in the kernel and a previous step in the path to remove
the strcpy function from the kernel entirely [1].

[1] https://github.com/KSPP/linux/issues/88

Signed-off-by: Len Baker <len.baker@gmx.com>
---
Hi,

This patch doesn't change the logic. I think it is better to use the
current logic and not use always the plus and minus signs as suggested
in the previous version. I don't like the idea that 0 has sign.

Thanks,
Len

The previous versions can be found in:

v1
https://lore.kernel.org/linux-hardening/20210801171157.17858-1-len.baker@gmx.com/

v2
https://lore.kernel.org/linux-hardening/20210807152225.9403-1-len.baker@gmx.com/

Changelog v1 -> v2
- Modify the commit changelog to inform that the motivation of this
  patch is to remove the strcpy() function from the kernel entirely
  (Jonathan Cameron).

Changelog v2 -> v3
- Rewrite the code using devm_kstrdup() and devm_kasprintf() functions
  (Andy Shevchenko).
- Rebase against v5.14-rc5.

 drivers/iio/imu/inv_mpu6050/inv_mpu_magn.c | 32 ++++++++++++++--------
 1 file changed, 20 insertions(+), 12 deletions(-)

diff --git a/drivers/iio/imu/inv_mpu6050/inv_mpu_magn.c b/drivers/iio/imu/inv_mpu6050/inv_mpu_magn.c
index f282e9cc34c5..9c359f894631 100644
--- a/drivers/iio/imu/inv_mpu6050/inv_mpu_magn.c
+++ b/drivers/iio/imu/inv_mpu6050/inv_mpu_magn.c
@@ -264,6 +264,7 @@ int inv_mpu_magn_set_orient(struct inv_mpu6050_state *st)
 	const char *orient;
 	char *str;
 	int i;
+	struct device *dev;

 	/* fill magnetometer orientation */
 	switch (st->chip_type) {
@@ -279,21 +280,28 @@ int inv_mpu_magn_set_orient(struct inv_mpu6050_state *st)
 		st->magn_orient.rotation[4] = st->orientation.rotation[1];
 		st->magn_orient.rotation[5] = st->orientation.rotation[2];
 		/* z <- -z */
+		dev = regmap_get_device(st->map);
 		for (i = 0; i < 3; ++i) {
 			orient = st->orientation.rotation[6 + i];
-			/* use length + 2 for adding minus sign if needed */
-			str = devm_kzalloc(regmap_get_device(st->map),
-					   strlen(orient) + 2, GFP_KERNEL);
-			if (str == NULL)
+
+			/*
+			 * The value is inverted according to the following
+			 * rules:
+			 *
+			 * 1) Drop leading minus.
+			 * 2) Add leading minus.
+			 * 3) Leave 0 as is.
+			 */
+			if (orient[0] == '-')
+				str = devm_kstrdup(dev, orient + 1, GFP_KERNEL);
+			else if (orient[0] != '0' || orient[1] != '\0')
+				str = devm_kasprintf(dev, GFP_KERNEL, "-%s", orient);
+			else
+				str = devm_kstrdup(dev, orient, GFP_KERNEL);
+
+			if (!str)
 				return -ENOMEM;
-			if (strcmp(orient, "0") == 0) {
-				strcpy(str, orient);
-			} else if (orient[0] == '-') {
-				strcpy(str, &orient[1]);
-			} else {
-				str[0] = '-';
-				strcpy(&str[1], orient);
-			}
+
 			st->magn_orient.rotation[6 + i] = str;
 		}
 		break;
--
2.25.1

Re: [PATCH v3] drivers/iio: Remove all strcpy() uses

[Andy Shevchenko]

On Sat, Aug 14, 2021 at 12:06 PM Len Baker <len.baker@gmx.com> wrote:
>
> strcpy() performs no bounds checking on the destination buffer. This
> could result in linear overflows beyond the end of the buffer, leading
> to all kinds of misbehaviors. So, remove all the uses and add
> devm_kstrdup() or devm_kasprintf() instead.
>
> This patch is an effort to clean up the proliferation of str*()
> functions in the kernel and a previous step in the path to remove
> the strcpy function from the kernel entirely [1].
>
> [1] https://github.com/KSPP/linux/issues/88

Thanks for an update, my comments below.

...

> This patch doesn't change the logic. I think it is better to use the
> current logic and not use always the plus and minus signs as suggested
> in the previous version. I don't like the idea that 0 has sign.

Agree on that, the safest way to go with.

...

>         const char *orient;
>         char *str;
>         int i;
> +       struct device *dev;

Please, keep this in reversed xmas tree order (longer lines first).

...

> +               dev = regmap_get_device(st->map);

I haven't checked the code in between, but maybe it's possible to move
an assignment directly to the definition block above.

...

> +                       /*
> +                        * The value is inverted according to the following

"to one of the"
And technically speaking "inversion" is not the same as negation
(which is "sign inversion").

> +                        * rules:
> +                        *
> +                        * 1) Drop leading minus.
> +                        * 2) Add leading minus.
> +                        * 3) Leave 0 as is.
> +                        */
> +                       if (orient[0] == '-')
> +                               str = devm_kstrdup(dev, orient + 1, GFP_KERNEL);

> +                       else if (orient[0] != '0' || orient[1] != '\0')
> +                               str = devm_kasprintf(dev, GFP_KERNEL, "-%s", orient);

I would go with the logic I suggested later on, i.e.

                       else if (orient[0] == '0' && orient[1] == '\0')
                               str = devm_kstrdup(dev, orient, GFP_KERNEL);

and below changed accordingly. It will clarify the "0" check.

> +                       else
> +                               str = devm_kstrdup(dev, orient, GFP_KERNEL);

> +

Redundant blank line.

> +                       if (!str)
>                                 return -ENOMEM;

-- 
With Best Regards,
Andy Shevchenko