Re: [PATCH net] tcp: reject devmem tx with fastopen and repair

[Stanislav Fomichev]

On 10/05, Kaifeng Wang wrote:
> tcp_sendmsg_locked() enforces that devmem TX can only proceed if the
> zero-copy path is active and a valid dmabuf binding exists. However,
> subsequent branches in tcp_sendmsg_locked() can still intercept the
> message before it reaches the devmem zero-copy loop:
> 
> 1. TCP Fast Open (MSG_FASTOPEN or DEFER_CONNECT):
>    If TCP_FASTOPEN_CONNECT is set, the socket may have a valid dst with
>    NETIF_F_SG (so zc == MSG_ZEROCOPY and binding is present), but
>    tcp_sendmsg_fastopen() -> tcp_send_syn_data() will use
>    copy_page_from_iter() to byte-copy from the iterator. Since iov_base
>    represents dma-buf offsets rather than user virtual addresses, this
>    misinterprets offsets as user pointers and copies arbitrary user memory
>    into the SYN packet.
> 
> 2. TCP repair mode:
>    If tp->repair is enabled with TCP_RECV_QUEUE, tcp_send_rcvq() similarly
>    calls skb_copy_datagram_from_iter(), byte-copying from the iterator.
> 
> Neither path supports or makes sense for devmem transmission. Reject devmem
> sends if Fast Open or repair mode is active.
> 
> This pre-existing issue was identified by Sashiko AI review on commit
> 125755776bc6 ("tcp: reject non zerocopy devmem tx") and has not been
> hit in production.

Flagged by an AI review: we do tp->repair check before
sk_stream_wait_connect which drops/requires the socket lock. So
technically someone can setsockopt(tcp_repair) which the connection
handshake is happening. Sounds reasonable?