Re: [PATCH mtd-utils] misc-utils: Fix integer overflow in docfdisk.c

[Zhihao Cheng <chengzhihao1@huawei.com>]

在 2024/12/17 20:44, Anton Moryakov 写道:
> Report of the static analyzer:
> Possible integer underflow: left operand is tainted. An integer underflow may occur due to arithmetic operation (subtraction) between variable 'block' and value '1', when 'block' is tainted { [-2147483648, 2147483647] }
> 
> Corrections explained:
> Added the block <= 0 check, which prevents integer underflow when calculating block - 1 and ensures that block is always positive.
> 
> Triggers found by static analyzer Svace.
> 
> Signed-off-by: Anton Moryakov <ant.v.moryakov@gmail.com>
> 
> ---
>   misc-utils/docfdisk.c | 4 ++++
>   1 file changed, 4 insertions(+)
> 
> diff --git a/misc-utils/docfdisk.c b/misc-utils/docfdisk.c
> index 486ce29..e473ceb 100644
> --- a/misc-utils/docfdisk.c
> +++ b/misc-utils/docfdisk.c
> @@ -255,6 +255,10 @@ int main(int argc, char **argv)
>   		ip->firstUnit = cpu_to_le32(block);
>   		if (!nblocks[i])
>   			nblocks[i] = totblocks - block;
> +		if (block <= 0) {
> +			fprintf(stderr, "Error: Invalid block value (%d). Block must be greater than 0.\n", block);
> +			return -1;
> +		}

In the previous code, 'block' is already initialized as:
  block = mhoffs / unitsize;
  block++;
So I don't think underflow could happen.
>   		ip->virtualUnits = cpu_to_le32(nblocks[i]);
>   		block += nblocks[i];
>   		ip->lastUnit = cpu_to_le32(block-1);
> 


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/