netdev - Re: [PATCH] bpf: Don't redirect packets with pkt

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <CANn89iK2JH5q3WUH9G58-WVhkaxpCvLbWrgKe17nKyethTr9bw@mail.gmail.com>
Date:   Fri, 17 Jun 2022 10:10:38 +0200
From:   Eric Dumazet <edumazet@...gle.com>
To:     Di Zhu <zhudi2@...wei.com>
Cc:     Alexei Starovoitov <ast@...nel.org>,
        Daniel Borkmann <daniel@...earbox.net>,
        Andrii Nakryiko <andrii@...nel.org>,
        Yonghong Song <yhs@...com>, David Miller <davem@...emloft.net>,
        netdev <netdev@...r.kernel.org>, bpf <bpf@...r.kernel.org>,
        rose.chen@...wei.com,
        syzbot+7a12909485b94426aceb@...kaller.appspotmail.com
Subject: Re: [PATCH] bpf: Don't redirect packets with pkt_len 0

On Fri, Jun 17, 2022 at 9:19 AM Di Zhu <zhudi2@...wei.com> wrote:
>
> Syzbot found an issue [1]: fq_codel_drop() try to drop a flow whitout any
> skbs, that is, the flow->head is null.
> The root cause, as the [2] says, is because that bpf_prog_test_run_skb()
> run a bpf prog which redirects empty skbs.
> So we should determine whether the length of the packet modified by bpf
> prog or others like bpf_prog_test is 0 before forwarding it directly.
>
> LINK: [1] https://syzkaller.appspot.com/bug?id=0b84da80c2917757915afa89f7738a9d16ec96c5
> LINK: [2] https://www.spinics.net/lists/netdev/msg777503.html
>
> Reported-by: syzbot+7a12909485b94426aceb@...kaller.appspotmail.com
> Signed-off-by: Di Zhu <zhudi2@...wei.com>
> ---
>  net/core/filter.c | 3 +++
>  1 file changed, 3 insertions(+)
>
> diff --git a/net/core/filter.c b/net/core/filter.c
> index 5af58eb48587..c7fbfa90898a 100644
> --- a/net/core/filter.c
> +++ b/net/core/filter.c
> @@ -2156,6 +2156,9 @@ static int __bpf_redirect_common(struct sk_buff *skb, struct net_device *dev,
>  static int __bpf_redirect(struct sk_buff *skb, struct net_device *dev,
>                           u32 flags)
>  {
> +       if (unlikely(skb->len == 0))
> +               return -EINVAL;
> +

You focus again on fq_codel, but we have a more generic issue at hand.

I said that most drivers will assume packets are Ethernet ones, having
at least an Ethernet header in them.

Also returning -EINVAL will leak the skb :/

I think a better fix would be to make sure the skb carries an expected
packet length,
and this probably differs in __bpf_redirect_common() and
__bpf_redirect_no_mac() ?

Current test in __bpf_redirect_common() seems not good enough.

+       /* Verify that a link layer header is carried */
+       if (unlikely(skb->mac_header >= skb->network_header)) {
+               kfree_skb(skb);
+               return -ERANGE;
+       }
+

It should check that the link layer header size is >= dev->min_header_len


>         if (dev_is_mac_header_xmit(dev))
>                 return __bpf_redirect_common(skb, dev, flags);
>         else
> --
> 2.27.0
>