Re: [PATCH bpf] bpf,tcp: avoid infinite recursion in BPF_SOCK_OPS_HDR_OPT_LEN_CB

[Alexei Starovoitov]

On Tue, Apr 14, 2026 at 3:57 AM Jiayuan Chen <jiayuan.chen@xxxxxxxxx> wrote:
>
> A BPF_PROG_TYPE_SOCK_OPS program can set BPF_SOCK_OPS_WRITE_HDR_OPT_CB_FLAG
> to inject custom TCP header options. When the kernel builds a TCP packet,
> it calls tcp_established_options() to calculate the header size, which
> invokes bpf_skops_hdr_opt_len() to trigger the BPF_SOCK_OPS_HDR_OPT_LEN_CB
> callback.
>
> If the BPF program calls bpf_setsockopt(TCP_NODELAY) inside this callback,
> __tcp_sock_set_nodelay() will call tcp_push_pending_frames(), which calls
> tcp_current_mss(), which calls tcp_established_options() again,
> re-triggering the same BPF callback. This creates an infinite recursion
> that exhausts the kernel stack and causes a panic.
>
> BPF_SOCK_OPS_HDR_OPT_LEN_CB
>   -> bpf_setsockopt(TCP_NODELAY)
>         -> tcp_push_pending_frames()
>           -> tcp_current_mss()
>                 -> tcp_established_options()
>                   -> bpf_skops_hdr_opt_len()
>                            /* infinite recursion */
>                         -> BPF_SOCK_OPS_HDR_OPT_LEN_CB
>
> A similar reentrancy issue exists for TCP congestion control, which is
> guarded by tp->bpf_chg_cc_inprogress. Adopt the same approach: introduce
> tp->bpf_hdr_opt_len_cb_inprogress, set it before invoking the callback in
> bpf_skops_hdr_opt_len(), and check it in sol_tcp_sockopt() to reject
> bpf_setsockopt(TCP_NODELAY) calls that would trigger
> tcp_push_pending_frames() and cause the recursion.
>
> Reported-by: Quan Sun <2022090917019@xxxxxxxxxxxxxxxx>
> Reported-by: Yinhao Hu <dddddd@xxxxxxxxxxx>
> Reported-by: Kaiyan Mei <M202472210@xxxxxxxxxxx>
> Reported-by: Dongliang Mu <dzm91@xxxxxxxxxxx>
> Closes: https://lore.kernel.org/bpf/d1d523c9-6901-4454-a183-94462b8f3e4e@xxxxxxxxxxxxxxxx/
> Fixes: 0813a841566f ("bpf: tcp: Allow bpf prog to write and parse TCP header option")
> Signed-off-by: Jiayuan Chen <jiayuan.chen@xxxxxxxxx>
> ---
>  Documentation/networking/net_cachelines/tcp_sock.rst |  1 +
>  include/linux/tcp.h                                  | 11 ++++++++++-
>  net/core/filter.c                                    |  4 ++++
>  net/ipv4/tcp_minisocks.c                             |  1 +
>  net/ipv4/tcp_output.c                                |  3 +++
>  5 files changed, 19 insertions(+), 1 deletion(-)
>
> diff --git a/Documentation/networking/net_cachelines/tcp_sock.rst b/Documentation/networking/net_cachelines/tcp_sock.rst
> index 563daea10d6c..07d3226d90cc 100644
> --- a/Documentation/networking/net_cachelines/tcp_sock.rst
> +++ b/Documentation/networking/net_cachelines/tcp_sock.rst
> @@ -152,6 +152,7 @@ unsigned_int                  keepalive_intvl
>  int                           linger2
>  u8                            bpf_sock_ops_cb_flags
>  u8:1                          bpf_chg_cc_inprogress
> +u8:1                          bpf_hdr_opt_len_cb_inprogress
>  u16                           timeout_rehash
>  u32                           rcv_ooopack
>  u32                           rcv_rtt_last_tsecr
> diff --git a/include/linux/tcp.h b/include/linux/tcp.h
> index f72eef31fa23..2bfb73cf922e 100644
> --- a/include/linux/tcp.h
> +++ b/include/linux/tcp.h
> @@ -475,12 +475,21 @@ struct tcp_sock {
>         u8      bpf_sock_ops_cb_flags;  /* Control calling BPF programs
>                                          * values defined in uapi/linux/tcp.h
>                                          */
> -       u8      bpf_chg_cc_inprogress:1; /* In the middle of
> +       u8      bpf_chg_cc_inprogress:1, /* In the middle of
>                                           * bpf_setsockopt(TCP_CONGESTION),
>                                           * it is to avoid the bpf_tcp_cc->init()
>                                           * to recur itself by calling
>                                           * bpf_setsockopt(TCP_CONGESTION, "itself").
>                                           */
> +               bpf_hdr_opt_len_cb_inprogress:1; /* It is set before invoking the
> +                                                 * callback so that a nested
> +                                                 * bpf_setsockopt(TCP_NODELAY) or
> +                                                 * bpf_setsockopt(TCP_CORK) cannot
> +                                                 * trigger tcp_push_pending_frames(),
> +                                                 * which would call tcp_current_mss()
> +                                                 * -> bpf_skops_hdr_opt_len(), causing
> +                                                 * infinite recursion.

Let's not add new bits.
Reuse existing and test/check all in one place,
like commit 061ff040710e9 did.

pw-bot: cr