| Message ID | 20240224-tcp-ao-tracepoints-v1-0-15f31b7f30a7@arista.com |
|---|---|
| Headers | show |
| Series | net/tcp: TCP-AO and TCP-MD5 tracepoints | expand |
On 2/24/24 09:30, Eric Dumazet wrote: > On Sat, Feb 24, 2024 at 10:04 AM Dmitry Safonov <dima@arista.com> wrote: >> >> Two reasons: >> 1. It's grown up enough >> 2. In order to not do header spaghetti by including >> <trace/events/tcp.h>, which is necessary for TCP tracepoints. >> >> Signed-off-by: Dmitry Safonov <dima@arista.com> > > Okay, but what about CONFIG_IPV6=m ? > > I do not see any EXPORT_SYMBOL() in this patch. Ouch. I keep forgetting about that case, will fix in v2. Thanks, Dmitry
On Sat, Feb 24, 2024 at 09:04:11AM +0000, Dmitry Safonov wrote: > Two reasons: > 1. It's grown up enough > 2. In order to not do header spaghetti by including > <trace/events/tcp.h>, which is necessary for TCP tracepoints. > > Signed-off-by: Dmitry Safonov <dima@arista.com> ... > diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c > index c82dc42f57c6..5fd61ae6bcc9 100644 > --- a/net/ipv4/tcp.c > +++ b/net/ipv4/tcp.c > @@ -4485,6 +4485,78 @@ EXPORT_SYMBOL(tcp_inbound_md5_hash); > > #endif > > +/* Called with rcu_read_lock() */ > +enum skb_drop_reason > +tcp_inbound_hash(struct sock *sk, const struct request_sock *req, > + const struct sk_buff *skb, > + const void *saddr, const void *daddr, > + int family, int dif, int sdif) > +{ > + const struct tcphdr *th = tcp_hdr(skb); > + const struct tcp_ao_hdr *aoh; > + const __u8 *md5_location; > + int l3index; > + > + /* Invalid option or two times meet any of auth options */ > + if (tcp_parse_auth_options(th, &md5_location, &aoh)) { > + tcp_hash_fail("TCP segment has incorrect auth options set", > + family, skb, ""); > + return SKB_DROP_REASON_TCP_AUTH_HDR; > + } > + > + if (req) { > + if (tcp_rsk_used_ao(req) != !!aoh) { > + u8 keyid, rnext, maclen; > + > + if (aoh) { > + keyid = aoh->keyid; > + rnext = aoh->rnext_keyid; > + maclen = tcp_ao_hdr_maclen(aoh); > + } else { > + keyid = rnext = maclen = 0; > + } Hi Dmitry, it looks like keyid is set but otherwise unused. Flagged by W=1 builds with gcc-13 and clang-17. > + > + NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPAOBAD); > + tcp_hash_fail("TCP connection can't start/end using TCP-AO", > + family, skb, "%s", > + !aoh ? "missing AO" : "AO signed"); > + return SKB_DROP_REASON_TCP_AOFAILURE; > + } > + } > + > + /* sdif set, means packet ingressed via a device > + * in an L3 domain and dif is set to the l3mdev > + */ > + l3index = sdif ? dif : 0; > + > + /* Fast path: unsigned segments */ > + if (likely(!md5_location && !aoh)) { > + /* Drop if there's TCP-MD5 or TCP-AO key with any rcvid/sndid > + * for the remote peer. On TCP-AO established connection > + * the last key is impossible to remove, so there's > + * always at least one current_key. > + */ > + if (tcp_ao_required(sk, saddr, family, l3index, true)) { > + tcp_hash_fail("AO hash is required, but not found", > + family, skb, "L3 index %d", l3index); > + return SKB_DROP_REASON_TCP_AONOTFOUND; > + } > + if (unlikely(tcp_md5_do_lookup(sk, l3index, saddr, family))) { > + NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPMD5NOTFOUND); > + tcp_hash_fail("MD5 Hash not found", > + family, skb, "L3 index %d", l3index); > + return SKB_DROP_REASON_TCP_MD5NOTFOUND; > + } > + return SKB_NOT_DROPPED_YET; > + } > + > + if (aoh) > + return tcp_inbound_ao_hash(sk, skb, family, req, l3index, aoh); > + > + return tcp_inbound_md5_hash(sk, skb, saddr, daddr, family, > + l3index, md5_location); > +} > + > void tcp_done(struct sock *sk) > { > struct request_sock *req; > > -- > 2.43.0 > >
On Mon, Feb 26, 2024 at 8:43 PM Simon Horman <horms@kernel.org> wrote: > > On Sat, Feb 24, 2024 at 09:04:11AM +0000, Dmitry Safonov wrote: [..] > > + if (req) { > > + if (tcp_rsk_used_ao(req) != !!aoh) { > > + u8 keyid, rnext, maclen; > > + > > + if (aoh) { > > + keyid = aoh->keyid; > > + rnext = aoh->rnext_keyid; > > + maclen = tcp_ao_hdr_maclen(aoh); > > + } else { > > + keyid = rnext = maclen = 0; > > + } > > Hi Dmitry, > > it looks like keyid is set but otherwise unused. > > Flagged by W=1 builds with gcc-13 and clang-17. Hi Simon, Yeah, I think I didn't notice it when I was splitting the WIP patch. It should be in the very next patch that uses them: + trace_tcp_ao_handshake_failure(sk, skb, keyid, rnext, maclen); Thanks for the report, going to fix in v2, Dmitry
I tried to split it a bit, maybe I could even go further and split by TRACE_EVENT_CLASS() changes, but not sure if it adds any value. But at least all preparation patches are separate. I wasn't sure if I should just remove tcp_hash_fail() as I did in this version, or rather put it under CONFIG_TCP_..., making it disabled by default and with a warning of deprecated, scheduled for removal. Maybe this won't cause any problems for anybody and I'm just too cautious of breaking others. Anyways, version 1, thanks for any reviews! Signed-off-by: Dmitry Safonov <dima@arista.com> --- Dmitry Safonov (10): net/tcp: Use static_branch_tcp_{md5,ao} to drop ifdefs net/tcp: Add a helper tcp_ao_hdr_maclen() net/tcp: Move tcp_inbound_hash() from headers net/tcp: Add tcp-md5 and tcp-ao tracepoints net/tcp: Remove tcp_hash_fail() selftests/net: Clean-up double assignment selftests/net: Provide test_snprintf() helper selftests/net: Be consistnat in kconfig checks selftests/net: Don't forget to close nsfd after switch_save_ns() selftest/net: Add trace events matching to tcp_ao include/net/tcp.h | 79 +- include/net/tcp_ao.h | 42 +- include/trace/events/tcp.h | 317 ++++++++ net/ipv4/tcp.c | 86 ++- net/ipv4/tcp_ao.c | 24 +- net/ipv4/tcp_input.c | 8 +- net/ipv4/tcp_ipv4.c | 8 +- net/ipv4/tcp_output.c | 2 + tools/testing/selftests/net/tcp_ao/Makefile | 2 +- tools/testing/selftests/net/tcp_ao/bench-lookups.c | 2 +- tools/testing/selftests/net/tcp_ao/connect-deny.c | 18 +- tools/testing/selftests/net/tcp_ao/connect.c | 2 +- tools/testing/selftests/net/tcp_ao/icmps-discard.c | 2 +- .../testing/selftests/net/tcp_ao/key-management.c | 18 +- tools/testing/selftests/net/tcp_ao/lib/aolib.h | 150 +++- tools/testing/selftests/net/tcp_ao/lib/ftrace.c | 846 +++++++++++++++++++++ tools/testing/selftests/net/tcp_ao/lib/kconfig.c | 31 +- tools/testing/selftests/net/tcp_ao/lib/setup.c | 15 +- tools/testing/selftests/net/tcp_ao/lib/sock.c | 1 - tools/testing/selftests/net/tcp_ao/lib/utils.c | 26 + tools/testing/selftests/net/tcp_ao/restore.c | 18 +- tools/testing/selftests/net/tcp_ao/rst.c | 2 +- tools/testing/selftests/net/tcp_ao/self-connect.c | 19 +- tools/testing/selftests/net/tcp_ao/seq-ext.c | 10 +- .../selftests/net/tcp_ao/setsockopt-closed.c | 2 +- tools/testing/selftests/net/tcp_ao/unsigned-md5.c | 28 +- 26 files changed, 1576 insertions(+), 182 deletions(-) --- base-commit: d662c5b3ce6dbed9d0991bc83001bbcc4a9bc2f8 change-id: 20240224-tcp-ao-tracepoints-0ea8ba11467a Best regards,