On Wed, Jun 20, 2018 at 11:26 AM Kees Cook <keesc...@chromium.org> wrote:
>
> In the quest to remove all stack VLA usage from the kernel[1], this
> uses the maximum allocation size for the stack and adds a sanity check,
> similar to what has already be done for the regular rave-sp driver.
>
> [1]
> https://lkml.kernel.org/r/CA+55aFzCG-zNmZwX4A2FQpadafLfEzK6CC=qpxydaacu1rq...@mail.gmail.com
>
> Signed-off-by: Kees Cook <keesc...@chromium.org>
I completely forgot that I was using VLAs in this driver. Thanks for
fixing that!
Reviewed-by: Andrey Smirnov <andrew.smir...@gmail.com>
Tested-by: Andrey Smirnov <andrew.smir...@gmail.com>
> ---
> drivers/nvmem/rave-sp-eeprom.c | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/nvmem/rave-sp-eeprom.c b/drivers/nvmem/rave-sp-eeprom.c
> index 50aeea6ec6cc..66699d44f73d 100644
> --- a/drivers/nvmem/rave-sp-eeprom.c
> +++ b/drivers/nvmem/rave-sp-eeprom.c
> @@ -35,6 +35,7 @@ enum rave_sp_eeprom_header_size {
> RAVE_SP_EEPROM_HEADER_SMALL = 4U,
> RAVE_SP_EEPROM_HEADER_BIG = 5U,
> };
> +#define RAVE_SP_EEPROM_HEADER_MAX RAVE_SP_EEPROM_HEADER_BIG
>
> #define RAVE_SP_EEPROM_PAGE_SIZE 32U
>
> @@ -97,9 +98,12 @@ static int rave_sp_eeprom_io(struct rave_sp_eeprom *eeprom,
> const unsigned int rsp_size =
> is_write ? sizeof(*page) - sizeof(page->data) : sizeof(*page);
> unsigned int offset = 0;
> - u8 cmd[cmd_size];
> + u8 cmd[RAVE_SP_EEPROM_HEADER_MAX + sizeof(page->data)];
> int ret;
>
> + if (WARN_ON(cmd_size > sizeof(cmd)))
> + return -EINVAL;
> +
> cmd[offset++] = eeprom->address;
> cmd[offset++] = 0;
> cmd[offset++] = type;
> --
> 2.17.1
>
>
> --
> Kees Cook
> Pixel Security
