Linux ACPI
 help / color / mirror / Atom feed
From: Mika Westerberg <mika.westerberg@linux.intel.com>
To: Marco Scardovi <scardracs@disroot.org>
Cc: Mika Westerberg <westeri@kernel.org>,
	Andy Shevchenko <andriy.shevchenko@linux.intel.com>,
	Linus Walleij <linusw@kernel.org>,
	Bartosz Golaszewski <brgl@kernel.org>,
	linux-gpio@vger.kernel.org, linux-acpi@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH 1/2] gpiolib: acpi: prevent address truncation in OperationRegion handler
Date: Mon, 1 Jun 2026 07:02:38 +0200	[thread overview]
Message-ID: <20260601050238.GM3102@black.igk.intel.com> (raw)
In-Reply-To: <20260530094012.184276-2-scardracs@disroot.org>

Hi,

On Sat, May 30, 2026 at 11:40:11AM +0200, Marco Scardovi wrote:
> The ACPI address space handler for GPIO OperationRegions receives the
> pin offset as a 64-bit acpi_physical_address. However, the handler
> truncates this address to a u16 pin_index before validating it.
> 
> If an ACPI table attempts to access a pin offset greater than 65535,
> the truncation wraps the index around. This may result in accesses to
> unintended GPIO pins.

If you look at the ACPI spec:

https://uefi.org/specs/ACPI/6.5/06_Device_Configuration.html#connection-descriptors

the pin number is 2 bytes and 0xffff is defined as no connection. So the
firmware cannot really think that it can access GPIO outside of that range.

> Fix this by adding an explicit check to verify that the 64-bit address
> is less than agpio->pin_table_length before assigning it to the u16
> pin_index, returning AE_BAD_PARAMETER if it is out of bounds.
> Additionally, make the length calculation overflow-safe and change the
> types of length and loop counter to unsigned.
> 
> Signed-off-by: Marco Scardovi <scardracs@disroot.org>
> ---
>  drivers/gpio/gpiolib-acpi-core.c | 17 +++++++++++++----
>  1 file changed, 13 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/gpio/gpiolib-acpi-core.c b/drivers/gpio/gpiolib-acpi-core.c
> index eb8a40cfb7a9..049e4cbc14ed 100644
> --- a/drivers/gpio/gpiolib-acpi-core.c
> +++ b/drivers/gpio/gpiolib-acpi-core.c
> @@ -1087,10 +1087,10 @@ acpi_gpio_adr_space_handler(u32 function, acpi_physical_address address,
>  	struct gpio_chip *chip = achip->chip;
>  	struct acpi_resource_gpio *agpio;
>  	struct acpi_resource *ares;
> -	u16 pin_index = address;
> +	unsigned int length;
>  	acpi_status status;
> -	int length;
> -	int i;
> +	unsigned int i;
> +	u16 pin_index;
>  
>  	status = acpi_buffer_to_resource(achip->conn_info.connection,
>  					 achip->conn_info.length, &ares);
> @@ -1110,7 +1110,16 @@ acpi_gpio_adr_space_handler(u32 function, acpi_physical_address address,
>  		return AE_BAD_PARAMETER;
>  	}
>  
> -	length = min(agpio->pin_table_length, pin_index + bits);
> +	if (address >= agpio->pin_table_length) {
> +		ACPI_FREE(ares);
> +		return AE_BAD_PARAMETER;
> +	}
> +
> +	pin_index = address;
> +	if (bits > agpio->pin_table_length - pin_index)
> +		length = agpio->pin_table_length;
> +	else
> +		length = pin_index + bits;
>  	for (i = pin_index; i < length; ++i) {
>  		unsigned int pin = agpio->pin_table[i];
>  		struct acpi_gpio_connection *conn;
> -- 
> 2.54.0

  reply	other threads:[~2026-06-01  5:02 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-05-30  9:40 [PATCH 0/2] gpiolib: acpi: fix bounds-checking bugs in GPIO ACPI core Marco Scardovi
2026-05-30  9:40 ` [PATCH 1/2] gpiolib: acpi: prevent address truncation in OperationRegion handler Marco Scardovi
2026-06-01  5:02   ` Mika Westerberg [this message]
2026-06-01  6:31     ` Marco Scardovi
2026-05-30  9:40 ` [PATCH 2/2] gpiolib: acpi: fix out-of-bounds pointer arithmetic in acpi_gpio_package_count Marco Scardovi
2026-06-01  5:17   ` Mika Westerberg
2026-06-01  6:31     ` Marco Scardovi
2026-06-01  7:17       ` Mika Westerberg
2026-06-01  7:53         ` Marco Scardovi
2026-06-02  7:52 ` [PATCH 0/2] gpiolib: acpi: fix bounds-checking bugs in GPIO ACPI core Andy Shevchenko
2026-06-02  7:59   ` Marco Scardovi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260601050238.GM3102@black.igk.intel.com \
    --to=mika.westerberg@linux.intel.com \
    --cc=andriy.shevchenko@linux.intel.com \
    --cc=brgl@kernel.org \
    --cc=linusw@kernel.org \
    --cc=linux-acpi@vger.kernel.org \
    --cc=linux-gpio@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=scardracs@disroot.org \
    --cc=westeri@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox