Re: [PATCH 4.14 156/173] gpiolib: Respect error code of ->get_direction()

[Jon Hunter]

Hi Greg,

On 24/09/18 12:53, Greg Kroah-Hartman wrote:
> 4.14-stable review patch.  If anyone has any objections, please let me know.
> 
> ------------------
> 
> From: Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx>
> 
> [ Upstream commit 36b312792b97933dc07abe074f50941199bd357c ]
> 
> In case we try to lock GPIO pin as IRQ when something going wrong
> we print a misleading message.
> 
> Correct this by checking an error code from ->get_direction() in
> gpiochip_lock_as_irq() and printing a corresponding message.
> 
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx>
> Cc: Mika Westerberg <mika.westerberg@xxxxxxxxxxxxxxx>
> Signed-off-by: Linus Walleij <linus.walleij@xxxxxxxxxx>
> Signed-off-by: Sasha Levin <alexander.levin@xxxxxxxxxxxxx>
> Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
> ---
>  drivers/gpio/gpiolib.c |    6 ++++++
>  1 file changed, 6 insertions(+)
> 
> --- a/drivers/gpio/gpiolib.c
> +++ b/drivers/gpio/gpiolib.c
> @@ -2811,6 +2811,12 @@ int gpiochip_lock_as_irq(struct gpio_chi
>  	if (!chip->can_sleep && chip->get_direction) {
>  		int dir = chip->get_direction(chip, offset);
>  
> +		if (dir < 0) {
> +			chip_err(chip, "%s: cannot get GPIO direction\n",
> +				 __func__);
> +			return dir;
> +		}
> +
>  		if (dir)
>  			clear_bit(FLAG_IS_OUT, &desc->flags);
>  		else

This change is causing a boot regression on Tegra20 and Tegra30. After
this change was merged in mainline there was a fix for Tegra [0] that
was also needed.

On Tegra the above change causes the allocation of GPIOs to fail and
looks like this then exposes another bug somewhere else that finally
leads to a system crash.

If this is needed for stable, then we will need the fix for Tegra as well.

Cheers
Jon

[0] https://lkml.org/lkml/2018/7/17/871

-- 
nvpublic