On Mon, Dec 1, 2025 at 1:52 AM, Avnish Chouhan <[email protected]> wrote:
> Adding a check for invalid partition number. grub_strtoul() can fail
> in several scenarios like invalid input, overflow, etc will result in
> an invalid partition number which could lead to an undefined behavior.
>
> Signed-off-by: Avnish Chouhan <[email protected]>
> ---
> grub-core/kern/ieee1275/openfw.c | 10 ++++++++++
> 1 file changed, 10 insertions(+)
>
> diff --git a/grub-core/kern/ieee1275/openfw.c
> b/grub-core/kern/ieee1275/openfw.c
> index 3b492dd..29bdb72 100644
> --- a/grub-core/kern/ieee1275/openfw.c
> +++ b/grub-core/kern/ieee1275/openfw.c
> @@ -512,8 +512,18 @@ grub_ieee1275_encode_devname (const char *path)
> }
> if (partition && partition[0])
> {
> + grub_errno = GRUB_ERR_NONE;
> unsigned int partno = grub_strtoul (partition, 0, 0);
>
> + if (grub_errno != GRUB_ERR_NONE)
This check isn't reliable. Please see how grub_strtoul() is handled in commit
ac8a37dda (net/http: Allow use of non-standard TCP/IP ports). Even if it is
reverted, the code around grub_strtoul() is fully correct.
Alec Brown
> + {
> + grub_free (partition);
> + grub_free (device);
> + grub_free (encoding);
> + grub_error (GRUB_ERR_BAD_ARGUMENT, N_("invalid partition number"));
> + return NULL;
> + }
> +
> *optr++ = ',';
>
> if (grub_ieee1275_test_flag (GRUB_IEEE1275_FLAG_0_BASED_PARTITIONS))
> --
> 2.52.0
_______________________________________________
Grub-devel mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/grub-devel
