Re: [PATCH v2 bpf] bpf: devmap: fix stack-out-of-bounds write in get_upper_ifindexes()

["Toke Høiland-Jørgensen" <toke@redhat.com>]

Kohei Enju <kohei@enjuk.jp> writes:

> get_upper_ifindexes() iterates over all upper devices and writes their
> indices into an array without checking bounds.
>
> Also the callers assume that the max number of upper devices is
> MAX_NEST_DEV and allocate excluded_devices[1+MAX_NEST_DEV] on the stack,
> but that assumption is not correct and the number of upper devices could
> be larger than MAX_NEST_DEV (e.g., many macvlans), causing a
> stack-out-of-bounds write.
>
> Add a max parameter to get_upper_ifindexes() to avoid the issue.
>
> To reproduce, create more than MAX_NEST_DEV(8) macvlans on a device with
> an XDP program attached using BPF_F_BROADCAST | BPF_F_EXCLUDE_INGRESS.
> Then send a packet to the device to trigger the XDP redirect path.
>
> Reported-by: syzbot+10cc7f13760b31bd2e61@syzkaller.appspotmail.com
> Closes: https://lore.kernel.org/all/698c4ce3.050a0220.340abe.000b.GAE@google.com/T/
> Fixes: aeea1b86f936 ("bpf, devmap: Exclude XDP broadcast to master device")
> Signed-off-by: Kohei Enju <kohei@enjuk.jp>
> ---
> Changes:
>   v2:
>     - fix formatting, accounting that max-line-length is 100
>   v1: https://lore.kernel.org/bpf/20260216201428.65641-1-kohei@enjuk.jp/
> ---
>  kernel/bpf/devmap.c | 12 ++++++++----
>  1 file changed, 8 insertions(+), 4 deletions(-)
>
> diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
> index 2625601de76e..c8c49d86a403 100644
> --- a/kernel/bpf/devmap.c
> +++ b/kernel/bpf/devmap.c
> @@ -588,16 +588,18 @@ static inline bool is_ifindex_excluded(int *excluded, int num_excluded, int ifin
>  }
>  
>  /* Get ifindex of each upper device. 'indexes' must be able to hold at
> - * least MAX_NEST_DEV elements.
> + * least 'max' elements.
>   * Returns the number of ifindexes added.
>   */
> -static int get_upper_ifindexes(struct net_device *dev, int *indexes)
> +static int get_upper_ifindexes(struct net_device *dev, int *indexes, int max)
>  {
>  	struct net_device *upper;
>  	struct list_head *iter;
>  	int n = 0;
>  
>  	netdev_for_each_upper_dev_rcu(dev, upper, iter) {
> +		if (n >= max)
> +			break;

Hmm, how about returning an error here, and aborting the redirect
entirely? This silent truncation to an arbitrary number seems a bit
surprising, and very hard to debug. With an error, there will at least
be an error tracepoint to catch the failure.

-Toke