On Fri, 29 May 2020 16:06:59 -0700
John Fastabend <john.fastabend@xxxxxxxxx> wrote:
> KTLS uses a stream parser to collect TLS messages and send them to
> the upper layer tls receive handler. This ensures the tls receiver
> has a full TLS header to parse when it is run. However, when a
> socket has BPF_SK_SKB_STREAM_VERDICT program attached before KTLS
> is enabled we end up with two stream parsers running on the same
> socket.
>
> The result is both try to run on the same socket. First the KTLS
> stream parser runs and calls read_sock() which will tcp_read_sock
> which in turn calls tcp_rcv_skb(). This dequeues the skb from the
> sk_receive_queue. When this is done KTLS code then data_ready()
> callback which because we stacked KTLS on top of the bpf stream
> verdict program has been replaced with sk_psock_start_strp(). This
> will in turn kick the stream parser again and eventually do the
> same thing KTLS did above calling into tcp_rcv_skb() and dequeuing
> a skb from the sk_receive_queue.
>
> At this point the data stream is broke. Part of the stream was
> handled by the KTLS side some other bytes may have been handled
> by the BPF side. Generally this results in either missing data
> or more likely a "Bad Message" complaint from the kTLS receive
> handler as the BPF program steals some bytes meant to be in a
> TLS header and/or the TLS header length is no longer correct.
>
> We've already broke the idealized model where we can stack ULPs
> in any order with generic callbacks on the TX side to handle this.
> So in this patch we do the same thing but for RX side. We add
> a sk_psock_strp_enabled() helper so TLS can learn a BPF verdict
> program is running and add a tls_sw_has_ctx_rx() helper so BPF
> side can learn there is a TLS ULP on the socket.
>
> Then on BPF side we omit calling our stream parser to avoid
> breaking the data stream for the KTLS receiver. Then on the
> KTLS side we call BPF_SK_SKB_STREAM_VERDICT once the KTLS
> receiver is done with the packet but before it posts the
> msg to userspace. This gives us symmetry between the TX and
> RX halfs and IMO makes it usable again. On the TX side we
> process packets in this order BPF -> TLS -> TCP and on
> the receive side in the reverse order TCP -> TLS -> BPF.
>
> Discovered while testing OpenSSL 3.0 Alpha2.0 release.
>
> Fixes: d829e9c4112b5 ("tls: convert to generic sk_msg interface")
> Signed-off-by: John Fastabend <john.fastabend@xxxxxxxxx>
> ---
> include/linux/skmsg.h | 8 ++++++++
> include/net/tls.h | 9 +++++++++
> net/core/skmsg.c | 43 ++++++++++++++++++++++++++++++++++++++++---
> net/tls/tls_sw.c | 20 ++++++++++++++++++--
> 4 files changed, 75 insertions(+), 5 deletions(-)
>
> diff --git a/include/linux/skmsg.h b/include/linux/skmsg.h
> index ad31c9f..08674cd 100644
> --- a/include/linux/skmsg.h
> +++ b/include/linux/skmsg.h
> @@ -437,4 +437,12 @@ static inline void psock_progs_drop(struct sk_psock_progs *progs)
> psock_set_prog(&progs->skb_verdict, NULL);
> }
>
> +int sk_psock_tls_strp_read(struct sk_psock *psock, struct sk_buff *skb);
> +
> +static inline bool sk_psock_strp_enabled(struct sk_psock *psock)
> +{
> + if (!psock)
> + return false;
> + return psock->parser.enabled;
> +}
> #endif /* _LINUX_SKMSG_H */
> diff --git a/include/net/tls.h b/include/net/tls.h
> index bf9eb48..b74d59b 100644
> --- a/include/net/tls.h
> +++ b/include/net/tls.h
> @@ -567,6 +567,15 @@ static inline bool tls_sw_has_ctx_tx(const struct sock *sk)
> return !!tls_sw_ctx_tx(ctx);
> }
>
> +static inline bool tls_sw_has_ctx_rx(const struct sock *sk)
> +{
> + struct tls_context *ctx = tls_get_ctx(sk);
> +
> + if (!ctx)
> + return false;
> + return !!tls_sw_ctx_rx(ctx);
> +}
> +
> void tls_sw_write_space(struct sock *sk, struct tls_context *ctx);
> void tls_device_write_space(struct sock *sk, struct tls_context *ctx);
>
> diff --git a/net/core/skmsg.c b/net/core/skmsg.c
> index 9d72f71..351afbf 100644
> --- a/net/core/skmsg.c
> +++ b/net/core/skmsg.c
> @@ -7,6 +7,7 @@
>
> #include <net/sock.h>
> #include <net/tcp.h>
> +#include <net/tls.h>
>
> static bool sk_msg_try_coalesce_ok(struct sk_msg *msg, int elem_first_coalesce)
> {
> @@ -714,6 +715,38 @@ static void sk_psock_skb_redirect(struct sk_psock *psock, struct sk_buff *skb)
> }
> }
>
> +static void sk_psock_tls_verdict_apply(struct sk_psock *psock,
> + struct sk_buff *skb, int verdict)
> +{
> + switch (verdict) {
> + case __SK_REDIRECT:
> + sk_psock_skb_redirect(psock, skb);
> + break;
> + case __SK_PASS:
> + case __SK_DROP:
The two cases above need a "fallthrough;", right?
> + default:
> + break;
> + }
> +}
> +
> +int sk_psock_tls_strp_read(struct sk_psock *psock, struct sk_buff *skb)
> +{
> + struct bpf_prog *prog;
> + int ret = __SK_PASS;
> +
> + rcu_read_lock();
> + prog = READ_ONCE(psock->progs.skb_verdict);
> + if (likely(prog)) {
> + tcp_skb_bpf_redirect_clear(skb);
> + ret = sk_psock_bpf_run(psock, prog, skb);
> + ret = sk_psock_map_verd(ret, tcp_skb_bpf_redirect_fetch(skb));
> + }
> + rcu_read_unlock();
> + sk_psock_tls_verdict_apply(psock, skb, ret);
> + return ret;
> +}
> +EXPORT_SYMBOL_GPL(sk_psock_tls_strp_read);
> +
> static void sk_psock_verdict_apply(struct sk_psock *psock,
> struct sk_buff *skb, int verdict)
> {
> @@ -792,9 +825,13 @@ static void sk_psock_strp_data_ready(struct sock *sk)
> rcu_read_lock();
> psock = sk_psock(sk);
> if (likely(psock)) {
> - write_lock_bh(&sk->sk_callback_lock);
> - strp_data_ready(&psock->parser.strp);
> - write_unlock_bh(&sk->sk_callback_lock);
> + if (tls_sw_has_ctx_rx(sk)) {
> + psock->parser.saved_data_ready(sk);
> + } else {
> + write_lock_bh(&sk->sk_callback_lock);
> + strp_data_ready(&psock->parser.strp);
> + write_unlock_bh(&sk->sk_callback_lock);
> + }
> }
> rcu_read_unlock();
> }
> diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c
> index 2d399b6..61043c6 100644
> --- a/net/tls/tls_sw.c
> +++ b/net/tls/tls_sw.c
> @@ -1731,6 +1731,7 @@ int tls_sw_recvmsg(struct sock *sk,
> long timeo;
> bool is_kvec = iov_iter_is_kvec(&msg->msg_iter);
> bool is_peek = flags & MSG_PEEK;
> + bool bpf_strp_enabled;
> int num_async = 0;
>
> flags |= nonblock;
> @@ -1740,6 +1741,7 @@ int tls_sw_recvmsg(struct sock *sk,
>
> psock = sk_psock_get(sk);
> lock_sock(sk);
> + bpf_strp_enabled = sk_psock_strp_enabled(psock);
>
> /* Process pending decrypted records. It must be non-zero-copy */
> err = process_rx_list(ctx, msg, &control, &cmsg, 0, len, false,
> @@ -1793,11 +1795,12 @@ int tls_sw_recvmsg(struct sock *sk,
>
> if (to_decrypt <= len && !is_kvec && !is_peek &&
> ctx->control == TLS_RECORD_TYPE_DATA &&
> - prot->version != TLS_1_3_VERSION)
> + prot->version != TLS_1_3_VERSION &&
> + !sk_psock_strp_enabled(psock))
Is this recheck of parser state intentional? Or can we test for
"!bpf_strp_enabled" here also?
> zc = true;
>
> /* Do not use async mode if record is non-data */
> - if (ctx->control == TLS_RECORD_TYPE_DATA)
> + if (ctx->control == TLS_RECORD_TYPE_DATA && !bpf_strp_enabled)
> async_capable = ctx->async_capable;
> else
> async_capable = false;
> @@ -1847,6 +1850,19 @@ int tls_sw_recvmsg(struct sock *sk,
> goto pick_next_record;
>
> if (!zc) {
> + if (bpf_strp_enabled) {
> + err = sk_psock_tls_strp_read(psock, skb);
> + if (err != __SK_PASS) {
> + rxm->offset = rxm->offset + rxm->full_len;
> + rxm->full_len = 0;
> + if (err == __SK_DROP)
> + consume_skb(skb);
> + ctx->recv_pkt = NULL;
> + __strp_unpause(&ctx->strp);
> + continue;
> + }
> + }
> +
> if (rxm->full_len > len) {
> retain_skb = true;
> chunk = len;
>
