[PATCH 1/2] gpiolib: Fix NULL pointer deference

[PATCH 1/2] gpiolib: Fix NULL pointer deference

[Ricardo Ribalda Delgado]

Under some circumstances, a gpiochip might be half cleaned from the
gpio_device list.

This patch makes sure that the chip pointer is still valid, before
calling the match function.

[  104.088296] BUG: unable to handle kernel NULL pointer dereference at
0000000000000090
[  104.089772] IP: [<ffffffff813d2045>] of_gpiochip_find_and_xlate+0x15/0x80
[  104.128273] Call Trace:
[  104.129802]  [<ffffffff813d2030>] ? of_parse_own_gpio+0x1f0/0x1f0
[  104.131353]  [<ffffffff813cd910>] gpiochip_find+0x60/0x90
[  104.132868]  [<ffffffff813d21ba>] of_get_named_gpiod_flags+0x9a/0x120
...
[  104.141586]  [<ffffffff8163d12b>] gpio_led_probe+0x11b/0x360

Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>
---

We might want to cc: stable on these two patches

 drivers/gpio/gpiolib.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c
index d407f904a31c..0626cc96744b 100644
--- a/drivers/gpio/gpiolib.c
+++ b/drivers/gpio/gpiolib.c
@@ -858,7 +858,7 @@ struct gpio_chip *gpiochip_find(void *data,
 
 	spin_lock_irqsave(&gpio_lock, flags);
 	list_for_each_entry(gdev, &gpio_devices, list)
-		if (match(gdev->chip, data))
+		if (gdev->chip && match(gdev->chip, data))
 			break;
 
 	/* No match? */
-- 
2.8.1

[PATCH 2/2] gpiolib: Fix unaligned used of reference counters

[Ricardo Ribalda Delgado]

gpiolib relies on the reference counters to clean up the gpio_device
structure.

Although the number of get/put is properly aligned on gpiolib.c
itself, it does not take into consideration how the referece counters
are affected by other external functions such as cdev_add and device_add.

Because of this, after the last call to put_device, the reference counter
has a value of +3, therefore never calling gpiodevice_release.

Due to the fact that some of the device  has already been cleaned on
gpiochip_remove, the library will end up OOPsing the kernel (e.g. a call
to of_gpiochip_find_and_xlate).

Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>
---

We might want to cc: stable on these two patches

drivers/gpio/gpiolib.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c
index 0626cc96744b..d829f2127524 100644
--- a/drivers/gpio/gpiolib.c
+++ b/drivers/gpio/gpiolib.c
@@ -438,7 +438,6 @@ static void gpiodevice_release(struct device *dev)
 {
 	struct gpio_device *gdev = dev_get_drvdata(dev);
 
-	cdev_del(&gdev->chrdev);
 	list_del(&gdev->list);
 	ida_simple_remove(&gpio_ida, gdev->id);
 	kfree(gdev->label);
@@ -471,7 +470,6 @@ static int gpiochip_setup_dev(struct gpio_device *gdev)
 
 	/* From this point, the .release() function cleans up gpio_device */
 	gdev->dev.release = gpiodevice_release;
-	get_device(&gdev->dev);
 	pr_debug("%s: registered GPIOs %d to %d on device: %s (%s)\n",
 		 __func__, gdev->base, gdev->base + gdev->ngpio - 1,
 		 dev_name(&gdev->dev), gdev->chip->label ? : "generic");
@@ -759,6 +757,8 @@ void gpiochip_remove(struct gpio_chip *chip)
 	 * be removed, else it will be dangling until the last user is
 	 * gone.
 	 */
+	cdev_del(&gdev->chrdev);
+	device_del(&gdev->dev);
 	put_device(&gdev->dev);
 }
 EXPORT_SYMBOL_GPL(gpiochip_remove);
-- 
2.8.1

Re: [PATCH 1/2] gpiolib: Fix NULL pointer deference

[Linus Walleij]

On Fri, Jun 3, 2016 at 7:10 PM, Ricardo Ribalda Delgado
<ricardo.ribalda@gmail.com> wrote:

> Under some circumstances, a gpiochip might be half cleaned from the
> gpio_device list.
>
> This patch makes sure that the chip pointer is still valid, before
> calling the match function.
>
> [  104.088296] BUG: unable to handle kernel NULL pointer dereference at
> 0000000000000090
> [  104.089772] IP: [<ffffffff813d2045>] of_gpiochip_find_and_xlate+0x15/0x80
> [  104.128273] Call Trace:
> [  104.129802]  [<ffffffff813d2030>] ? of_parse_own_gpio+0x1f0/0x1f0
> [  104.131353]  [<ffffffff813cd910>] gpiochip_find+0x60/0x90
> [  104.132868]  [<ffffffff813d21ba>] of_get_named_gpiod_flags+0x9a/0x120
> ...
> [  104.141586]  [<ffffffff8163d12b>] gpio_led_probe+0x11b/0x360
>
> Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>

Oh there is the fix that Roger and Grygorii needs.

Applied to fixes and tagged for stable.

Sorry for screwing things up :(

Yours,
Linus Walleij

Re: [PATCH 2/2] gpiolib: Fix unaligned used of reference counters

[Linus Walleij]

On Fri, Jun 3, 2016 at 7:10 PM, Ricardo Ribalda Delgado
<ricardo.ribalda@gmail.com> wrote:

> gpiolib relies on the reference counters to clean up the gpio_device
> structure.
>
> Although the number of get/put is properly aligned on gpiolib.c
> itself, it does not take into consideration how the referece counters
> are affected by other external functions such as cdev_add and device_add.
>
> Because of this, after the last call to put_device, the reference counter
> has a value of +3, therefore never calling gpiodevice_release.
>
> Due to the fact that some of the device  has already been cleaned on
> gpiochip_remove, the library will end up OOPsing the kernel (e.g. a call
> to of_gpiochip_find_and_xlate).
>
> Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>

Again, thanks for fixing my stupid mistakes.

Applied and tagged for stable.

Yours,
Linus Walleij

Re: [PATCH 1/2] gpiolib: Fix NULL pointer deference

[Grygorii Strashko]

On 06/08/2016 11:39 AM, Linus Walleij wrote:
> On Fri, Jun 3, 2016 at 7:10 PM, Ricardo Ribalda Delgado
> <ricardo.ribalda@gmail.com> wrote:
>
>> Under some circumstances, a gpiochip might be half cleaned from the
>> gpio_device list.
>>
>> This patch makes sure that the chip pointer is still valid, before
>> calling the match function.
>>
>> [  104.088296] BUG: unable to handle kernel NULL pointer dereference at
>> 0000000000000090
>> [  104.089772] IP: [<ffffffff813d2045>] of_gpiochip_find_and_xlate+0x15/0x80
>> [  104.128273] Call Trace:
>> [  104.129802]  [<ffffffff813d2030>] ? of_parse_own_gpio+0x1f0/0x1f0
>> [  104.131353]  [<ffffffff813cd910>] gpiochip_find+0x60/0x90
>> [  104.132868]  [<ffffffff813d21ba>] of_get_named_gpiod_flags+0x9a/0x120
>> ...
>> [  104.141586]  [<ffffffff8163d12b>] gpio_led_probe+0x11b/0x360
>>
>> Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>
>
> Oh there is the fix that Roger and Grygorii needs.
>
> Applied to fixes and tagged for stable.
>
> Sorry for screwing things up :(
>

Np. :)

-- 
regards,
-grygorii