linux-kernel - Re: [PATCH] pinctrl: apple: Always return valid type in apple_gpio_irq

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20211101151836.GA6547@e124191.cambridge.arm.com>
Date:   Mon, 1 Nov 2021 15:18:36 +0000
From:   Joey Gouly <joey.gouly@....com>
To:     Sven Peter <sven@...npeter.dev>
Cc:     Hector Martin <marcan@...can.st>,
        Linus Walleij <linus.walleij@...aro.org>,
        Alyssa Rosenzweig <alyssa@...enzweig.io>,
        Marc Zyngier <maz@...nel.org>,
        linux-arm-kernel@...ts.infradead.org, linux-gpio@...r.kernel.org,
        linux-kernel@...r.kernel.org, nd@....com
Subject: Re: [PATCH] pinctrl: apple: Always return valid type in
 apple_gpio_irq_type

Hi Sven,

On Mon, Nov 01, 2021 at 04:06:40PM +0100, Sven Peter wrote:
> apple_gpio_irq_type can possibly return -EINVAL which triggers the
> following compile error with gcc 9 because the type no longer fits
> into the mask.
> 
>   drivers/pinctrl/pinctrl-apple-gpio.c: In function 'apple_gpio_irq_set_type':
>   ././include/linux/compiler_types.h:335:38: error: call to '__compiletime_assert_289' declared with attribute error: FIELD_PREP: value too large for the field
>     335 |  _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
>         |                                      ^
>   [...]
>   drivers/pinctrl/pinctrl-apple-gpio.c:294:7: note: in expansion of macro 'FIELD_PREP'
>     294 |       FIELD_PREP(REG_GPIOx_MODE, irqtype));
>         |       ^~~~~~~~~~
> 
> Fix this by making the return value always valid and instead checking
> for REG_GPIOx_IN_IRQ_OFF in apple_gpio_irq_set_type and return -EINVAL
> from there.
> 
> Fixes: a0f160ffcb83 ("pinctrl: add pinctrl/GPIO driver for Apple SoCs")
> Signed-off-by: Sven Peter <sven@...npeter.dev>
> ---
>  drivers/pinctrl/pinctrl-apple-gpio.c | 12 ++++++------
>  1 file changed, 6 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/pinctrl/pinctrl-apple-gpio.c b/drivers/pinctrl/pinctrl-apple-gpio.c
> index 0cc346bfc4c3..a7861079a650 100644
> --- a/drivers/pinctrl/pinctrl-apple-gpio.c
> +++ b/drivers/pinctrl/pinctrl-apple-gpio.c
> @@ -258,7 +258,7 @@ static void apple_gpio_irq_ack(struct irq_data *data)
>  	       pctl->base + REG_IRQ(irqgrp, data->hwirq));
>  }
>  
> -static int apple_gpio_irq_type(unsigned int type)
> +static unsigned int apple_gpio_irq_type(unsigned int type)
>  {
>  	switch (type & IRQ_TYPE_SENSE_MASK) {
>  	case IRQ_TYPE_EDGE_RISING:
> @@ -272,7 +272,7 @@ static int apple_gpio_irq_type(unsigned int type)
>  	case IRQ_TYPE_LEVEL_LOW:
>  		return REG_GPIOx_IN_IRQ_LO;
>  	default:
> -		return -EINVAL;
> +		return REG_GPIOx_IN_IRQ_OFF;
>  	}
>  }
>  
> @@ -288,7 +288,7 @@ static void apple_gpio_irq_unmask(struct irq_data *data)
>  {
>  	struct apple_gpio_pinctrl *pctl =
>  		gpiochip_get_data(irq_data_get_irq_chip_data(data));
> -	int irqtype = apple_gpio_irq_type(irqd_get_trigger_type(data));
> +	unsigned int irqtype = apple_gpio_irq_type(irqd_get_trigger_type(data));
>  
>  	apple_gpio_set_reg(pctl, data->hwirq, REG_GPIOx_MODE,
>  			   FIELD_PREP(REG_GPIOx_MODE, irqtype));
> @@ -313,10 +313,10 @@ static int apple_gpio_irq_set_type(struct irq_data *data,
>  {
>  	struct apple_gpio_pinctrl *pctl =
>  		gpiochip_get_data(irq_data_get_irq_chip_data(data));
> -	int irqtype = apple_gpio_irq_type(type);
> +	unsigned int irqtype = apple_gpio_irq_type(type);
>  
> -	if (irqtype < 0)
> -		return irqtype;
> +	if (irqtype == REG_GPIOx_IN_IRQ_OFF)
> +		return -EINVAL;
>  
>  	apple_gpio_set_reg(pctl, data->hwirq, REG_GPIOx_MODE,
>  			   FIELD_PREP(REG_GPIOx_MODE, irqtype));
> -- 
> 2.25.1
> 

This was previously reported, but was miscategorised it as a compiler bug, so thanks for fixing this!

Reviewed-by: Joey Gouly <joey.gouly@....com>

Thanks,
Joey