summaryrefslogtreecommitdiffstats
path: root/net/ipv4/tcp_ipv4.c
Commit message (Collapse)AuthorAgeFilesLines
* tcp: reduce accepted window in NEW_SYN_RECV stateEric Dumazet2024-05-271-6/+1
| | | | | | | | | | | | | | | | | | | | | | | | Jason commit made checks against ACK sequence less strict and can be exploited by attackers to establish spoofed flows with less probes. Innocent users might use tcp_rmem[1] == 1,000,000,000, or something more reasonable. An attacker can use a regular TCP connection to learn the server initial tp->rcv_wnd, and use it to optimize the attack. If we make sure that only the announced window (smaller than 65535) is used for ACK validation, we force an attacker to use 65537 packets to complete the 3WHS (assuming server ISN is unknown) Fixes: 378979e94e95 ("tcp: remove 64 KByte limit for initial tp->rcv_wnd value") Link: https://datatracker.ietf.org/meeting/119/materials/slides-119-tcpm-ghost-acks-00 Signed-off-by: Eric Dumazet <edumazet@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Reviewed-by: Jason Xing <kerneljasonxing@gmail.com> Link: https://lore.kernel.org/r/20240523130528.60376-1-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
* tcp: rstreason: handle timewait cases in the receive pathJason Xing2024-05-131-1/+1
| | | | | | | | | | There are two possible cases where TCP layer can send an RST. Since they happen in the same place, I think using one independent reason is enough to identify this special situation. Signed-off-by: Jason Xing <kernelxing@tencent.com> Link: https://lore.kernel.org/r/20240510122502.27850-5-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
* tcp: get rid of twsk_unique()Eric Dumazet2024-05-091-1/+0
| | | | | | | | | | | DCCP is going away soon, and had no twsk_unique() method. We can directly call tcp_twsk_unique() for TCP sockets. Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com> Link: https://lore.kernel.org/r/20240507164140.940547-1-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
* Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/netJakub Kicinski2024-05-091-1/+7
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | Cross-merge networking fixes after downstream PR. No conflicts. Adjacent changes: drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c 35d92abfbad8 ("net: hns3: fix kernel crash when devlink reload during initialization") 2a1a1a7b5fd7 ("net: hns3: add command queue trace for hns3") Signed-off-by: Jakub Kicinski <kuba@kernel.org>
| * tcp: Use refcount_inc_not_zero() in tcp_twsk_unique().Kuniyuki Iwashima2024-05-021-1/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Anderson Nascimento reported a use-after-free splat in tcp_twsk_unique() with nice analysis. Since commit ec94c2696f0b ("tcp/dccp: avoid one atomic operation for timewait hashdance"), inet_twsk_hashdance() sets TIME-WAIT socket's sk_refcnt after putting it into ehash and releasing the bucket lock. Thus, there is a small race window where other threads could try to reuse the port during connect() and call sock_hold() in tcp_twsk_unique() for the TIME-WAIT socket with zero refcnt. If that happens, the refcnt taken by tcp_twsk_unique() is overwritten and sock_put() will cause underflow, triggering a real use-after-free somewhere else. To avoid the use-after-free, we need to use refcount_inc_not_zero() in tcp_twsk_unique() and give up on reusing the port if it returns false. [0]: refcount_t: addition on 0; use-after-free. WARNING: CPU: 0 PID: 1039313 at lib/refcount.c:25 refcount_warn_saturate+0xe5/0x110 CPU: 0 PID: 1039313 Comm: trigger Not tainted 6.8.6-200.fc39.x86_64 #1 Hardware name: VMware, Inc. VMware20,1/440BX Desktop Reference Platform, BIOS VMW201.00V.21805430.B64.2305221830 05/22/2023 RIP: 0010:refcount_warn_saturate+0xe5/0x110 Code: 42 8e ff 0f 0b c3 cc cc cc cc 80 3d aa 13 ea 01 00 0f 85 5e ff ff ff 48 c7 c7 f8 8e b7 82 c6 05 96 13 ea 01 01 e8 7b 42 8e ff <0f> 0b c3 cc cc cc cc 48 c7 c7 50 8f b7 82 c6 05 7a 13 ea 01 01 e8 RSP: 0018:ffffc90006b43b60 EFLAGS: 00010282 RAX: 0000000000000000 RBX: ffff888009bb3ef0 RCX: 0000000000000027 RDX: ffff88807be218c8 RSI: 0000000000000001 RDI: ffff88807be218c0 RBP: 0000000000069d70 R08: 0000000000000000 R09: ffffc90006b439f0 R10: ffffc90006b439e8 R11: 0000000000000003 R12: ffff8880029ede84 R13: 0000000000004e20 R14: ffffffff84356dc0 R15: ffff888009bb3ef0 FS: 00007f62c10926c0(0000) GS:ffff88807be00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000020ccb000 CR3: 000000004628c005 CR4: 0000000000f70ef0 PKRU: 55555554 Call Trace: <TASK> ? refcount_warn_saturate+0xe5/0x110 ? __warn+0x81/0x130 ? refcount_warn_saturate+0xe5/0x110 ? report_bug+0x171/0x1a0 ? refcount_warn_saturate+0xe5/0x110 ? handle_bug+0x3c/0x80 ? exc_invalid_op+0x17/0x70 ? asm_exc_invalid_op+0x1a/0x20 ? refcount_warn_saturate+0xe5/0x110 tcp_twsk_unique+0x186/0x190 __inet_check_established+0x176/0x2d0 __inet_hash_connect+0x74/0x7d0 ? __pfx___inet_check_established+0x10/0x10 tcp_v4_connect+0x278/0x530 __inet_stream_connect+0x10f/0x3d0 inet_stream_connect+0x3a/0x60 __sys_connect+0xa8/0xd0 __x64_sys_connect+0x18/0x20 do_syscall_64+0x83/0x170 entry_SYSCALL_64_after_hwframe+0x78/0x80 RIP: 0033:0x7f62c11a885d Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d a3 45 0c 00 f7 d8 64 89 01 48 RSP: 002b:00007f62c1091e58 EFLAGS: 00000296 ORIG_RAX: 000000000000002a RAX: ffffffffffffffda RBX: 0000000020ccb004 RCX: 00007f62c11a885d RDX: 0000000000000010 RSI: 0000000020ccb000 RDI: 0000000000000003 RBP: 00007f62c1091e90 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000296 R12: 00007f62c10926c0 R13: ffffffffffffff88 R14: 0000000000000000 R15: 00007ffe237885b0 </TASK> Fixes: ec94c2696f0b ("tcp/dccp: avoid one atomic operation for timewait hashdance") Reported-by: Anderson Nascimento <anderson@allelesecurity.com> Closes: https://lore.kernel.org/netdev/37a477a6-d39e-486b-9577-3463f655a6b7@allelesecurity.com/ Suggested-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Link: https://lore.kernel.org/r/20240501213145.62261-1-kuniyu@amazon.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
* | rstreason: make it work in trace worldJason Xing2024-04-261-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | At last, we should let it work by introducing this reset reason in trace world. One of the possible expected outputs is: ... tcp_send_reset: skbaddr=xxx skaddr=xxx src=xxx dest=xxx state=TCP_ESTABLISHED reason=NOT_SPECIFIED Signed-off-by: Jason Xing <kernelxing@tencent.com> Reviewed-by: Steven Rostedt (Google) <rostedt@goodmis.org> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
* | tcp: support rstreason for passive resetJason Xing2024-04-261-4/+7
| | | | | | | | | | | | | | | | | | | | | | Reuse the dropreason logic to show the exact reason of tcp reset, so we can finally display the corresponding item in enum sk_reset_reason instead of reinventing new reset reasons. This patch replaces all the prior NOT_SPECIFIED reasons. Signed-off-by: Jason Xing <kernelxing@tencent.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
* | rstreason: prepare for passive resetJason Xing2024-04-261-5/+7
| | | | | | | | | | | | | | | | | | | | Adjust the parameter and support passing reason of reset which is for now NOT_SPECIFIED. No functional changes. Signed-off-by: Jason Xing <kernelxing@tencent.com> Acked-by: Matthieu Baerts (NGI0) <matttbe@kernel.org> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
* | tcp: avoid premature drops in tcp_add_backlog()Eric Dumazet2024-04-251-2/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | While testing TCP performance with latest trees, I saw suspect SOCKET_BACKLOG drops. tcp_add_backlog() computes its limit with : limit = (u32)READ_ONCE(sk->sk_rcvbuf) + (u32)(READ_ONCE(sk->sk_sndbuf) >> 1); limit += 64 * 1024; This does not take into account that sk->sk_backlog.len is reset only at the very end of __release_sock(). Both sk->sk_backlog.len and sk->sk_rmem_alloc could reach sk_rcvbuf in normal conditions. We should double sk->sk_rcvbuf contribution in the formula to absorb bubbles in the backlog, which happen more often for very fast flows. This change maintains decent protection against abuses. Fixes: c377411f2494 ("net: sk_add_backlog() take rmem_alloc into account") Signed-off-by: Eric Dumazet <edumazet@google.com> Link: https://lore.kernel.org/r/20240423125620.3309458-1-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
* | tcp: small optimization when TCP_TW_SYN is processedEric Dumazet2024-04-121-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When TCP_TW_SYN is processed, we perform a lookup to find a listener and jump back in tcp_v6_rcv() and tcp_v4_rcv() Paolo suggested that we do not have to check if the found socket is a TIME_WAIT or NEW_SYN_RECV one. Suggested-by: Paolo Abeni <pabeni@redhat.com> Link: https://lore.kernel.org/netdev/68085c8a84538cacaac991415e4ccc72f45e76c2.camel@redhat.com/ Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neal Cardwell <ncardwell@google.com> Link: https://lore.kernel.org/r/20240411082530.907113-1-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
* | tcp: replace TCP_SKB_CB(skb)->tcp_tw_isn with a per-cpu fieldEric Dumazet2024-04-091-2/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | TCP can transform a TIMEWAIT socket into a SYN_RECV one from a SYN packet, and the ISN of the SYNACK packet is normally generated using TIMEWAIT tw_snd_nxt : tcp_timewait_state_process() ... u32 isn = tcptw->tw_snd_nxt + 65535 + 2; if (isn == 0) isn++; TCP_SKB_CB(skb)->tcp_tw_isn = isn; return TCP_TW_SYN; This SYN packet also bypasses normal checks against listen queue being full or not. tcp_conn_request() ... __u32 isn = TCP_SKB_CB(skb)->tcp_tw_isn; ... /* TW buckets are converted to open requests without * limitations, they conserve resources and peer is * evidently real one. */ if ((syncookies == 2 || inet_csk_reqsk_queue_is_full(sk)) && !isn) { want_cookie = tcp_syn_flood_action(sk, rsk_ops->slab_name); if (!want_cookie) goto drop; } This was using TCP_SKB_CB(skb)->tcp_tw_isn field in skb. Unfortunately this field has been accidentally cleared after the call to tcp_timewait_state_process() returning TCP_TW_SYN. Using a field in TCP_SKB_CB(skb) for a temporary state is overkill. Switch instead to a per-cpu variable. As a bonus, we do not have to clear tcp_tw_isn in TCP receive fast path. It is temporarily set then cleared only in the TCP_TW_SYN dance. Fixes: 4ad19de8774e ("net: tcp6: fix double call of tcp_v6_fill_cb()") Fixes: eeea10b83a13 ("tcp: add tcp_v4_fill_cb()/tcp_v4_restore_cb()") Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
* | tcp: propagate tcp_tw_isn via an extra parameter to ->route_req()Eric Dumazet2024-04-091-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | tcp_v6_init_req() reads TCP_SKB_CB(skb)->tcp_tw_isn to find out if the request socket is created by a SYN hitting a TIMEWAIT socket. This has been buggy for a decade, lets directly pass the information from tcp_conn_request(). This is a preparatory patch to make the following one easier to review. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
* | net: skbuff: generalize the skb->decrypted bitJakub Kicinski2024-04-061-3/+1
| | | | | | | | | | | | | | | | | | | | The ->decrypted bit can be reused for other crypto protocols. Remove the direct dependency on TLS, add helpers to clean up the ifdefs leaking out everywhere. Signed-off-by: Jakub Kicinski <kuba@kernel.org> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* | trace: tcp: fully support trace_tcp_send_resetJason Xing2024-04-031-4/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Prior to this patch, what we can see by enabling trace_tcp_send is only happening under two circumstances: 1) active rst mode 2) non-active rst mode and based on the full socket That means the inconsistency occurs if we use tcpdump and trace simultaneously to see how rst happens. It's necessary that we should take into other cases into considerations, say: 1) time-wait socket 2) no socket ... By parsing the incoming skb and reversing its 4-tuple can we know the exact 'flow' which might not exist. Samples after applied this patch: 1. tcp_send_reset: skbaddr=XXX skaddr=XXX src=ip:port dest=ip:port state=TCP_ESTABLISHED 2. tcp_send_reset: skbaddr=000...000 skaddr=XXX src=ip:port dest=ip:port state=UNKNOWN Note: 1) UNKNOWN means we cannot extract the right information from skb. 2) skbaddr/skaddr could be 0 Signed-off-by: Jason Xing <kernelxing@tencent.com> Link: https://lore.kernel.org/r/20240401073605.37335-3-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
* | tcp/dccp: do not care about families in inet_twsk_purge()Eric Dumazet2024-04-011-1/+1
|/ | | | | | | | | | | | | | | | We lost ability to unload ipv6 module a long time ago. Instead of calling expensive inet_twsk_purge() twice, we can handle all families in one round. Also remove an extra line added in my prior patch, per Kuniyuki Iwashima feedback. Signed-off-by: Eric Dumazet <edumazet@google.com> Link: https://lore.kernel.org/netdev/20240327192934.6843-1-kuniyu@amazon.com/ Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com> Link: https://lore.kernel.org/r/20240329153203.345203-1-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
* tcp: make dropreason in tcp_child_process() workJason Xing2024-02-281-5/+7
| | | | | | | | | | It's time to let it work right now. We've already prepared for this:) Signed-off-by: Jason Xing <kernelxing@tencent.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Reviewed-by: David Ahern <dsahern@kernel.org> Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* tcp: make the dropreason really work when calling tcp_rcv_state_process()Jason Xing2024-02-281-1/+2
| | | | | | | | | | | Update three callers including both ipv4 and ipv6 and let the dropreason mechanism work in reality. Signed-off-by: Jason Xing <kernelxing@tencent.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Reviewed-by: David Ahern <dsahern@kernel.org> Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* tcp: directly drop skb in cookie check for ipv4Jason Xing2024-02-281-1/+1
| | | | | | | | | | | | Only move the skb drop from tcp_v4_do_rcv() to cookie_v4_check() itself, no other changes made. It can help us refine the specific drop reasons later. Signed-off-by: Jason Xing <kernelxing@tencent.com> Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* net/tcp: Consistently align TCP-AO option in the headerDmitry Safonov2023-12-061-2/+2
| | | | | | | | | | | | | | | Currently functions that pre-calculate TCP header options length use unaligned TCP-AO header + MAC-length for skb reservation. And the functions that actually write TCP-AO options into skb do align the header. Nothing good can come out of this for ((maclen % 4) != 0). Provide tcp_ao_len_aligned() helper and use it everywhere for TCP header options space calculations. Fixes: 1e03d32bea8e ("net/tcp: Add TCP-AO sign to outgoing packets") Signed-off-by: Dmitry Safonov <dima@arista.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
* net/tcp: Wire up l3index to TCP-AODmitry Safonov2023-10-271-3/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Similarly how TCP_MD5SIG_FLAG_IFINDEX works for TCP-MD5, TCP_AO_KEYF_IFINDEX is an AO-key flag that binds that MKT to a specified by L3 ifinndex. Similarly, without this flag the key will work in the default VRF l3index = 0 for connections. To prevent AO-keys from overlapping, it's restricted to add key B for a socket that has key A, which have the same sndid/rcvid and one of the following is true: - !(A.keyflags & TCP_AO_KEYF_IFINDEX) or !(B.keyflags & TCP_AO_KEYF_IFINDEX) so that any key is non-bound to a VRF - A.l3index == B.l3index both want to work for the same VRF Additionally, it's restricted to match TCP-MD5 keys for the same peer the following way: |--------------|--------------------|----------------|---------------| | | MD5 key without | MD5 key | MD5 key | | | l3index | l3index=0 | l3index=N | |--------------|--------------------|----------------|---------------| | TCP-AO key | | | | | without | reject | reject | reject | | l3index | | | | |--------------|--------------------|----------------|---------------| | TCP-AO key | | | | | l3index=0 | reject | reject | allow | |--------------|--------------------|----------------|---------------| | TCP-AO key | | | | | l3index=N | reject | allow | reject | |--------------|--------------------|----------------|---------------| This is done with the help of tcp_md5_do_lookup_any_l3index() to reject adding AO key without TCP_AO_KEYF_IFINDEX if there's TCP-MD5 in any VRF. This is important for case where sysctl_tcp_l3mdev_accept = 1 Similarly, for TCP-AO lookups tcp_ao_do_lookup() may be used with l3index < 0, so that __tcp_ao_key_cmp() will match TCP-AO key in any VRF. Signed-off-by: Dmitry Safonov <dima@arista.com> Acked-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* net/tcp: Add static_key for TCP-AODmitry Safonov2023-10-271-11/+14
| | | | | | | | | | | | | | | Similarly to TCP-MD5, add a static key to TCP-AO that is patched out when there are no keys on a machine and dynamically enabled with the first setsockopt(TCP_AO) adds a key on any socket. The static key is as well dynamically disabled later when the socket is destructed. The lifetime of enabled static key here is the same as ao_info: it is enabled on allocation, passed over from full socket to twsk and destructed when ao_info is scheduled for destruction. Signed-off-by: Dmitry Safonov <dima@arista.com> Acked-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* net/tcp: Ignore specific ICMPs for TCP-AO connectionsDmitry Safonov2023-10-271-0/+7
| | | | | | | | | | | | | | | | | | | | | | Similarly to IPsec, RFC5925 prescribes: ">> A TCP-AO implementation MUST default to ignore incoming ICMPv4 messages of Type 3 (destination unreachable), Codes 2-4 (protocol unreachable, port unreachable, and fragmentation needed -- ’hard errors’), and ICMPv6 Type 1 (destination unreachable), Code 1 (administratively prohibited) and Code 4 (port unreachable) intended for connections in synchronized states (ESTABLISHED, FIN-WAIT-1, FIN- WAIT-2, CLOSE-WAIT, CLOSING, LAST-ACK, TIME-WAIT) that match MKTs." A selftest (later in patch series) verifies that this attack is not possible in this TCP-AO implementation. Co-developed-by: Francesco Ruggeri <fruggeri@arista.com> Signed-off-by: Francesco Ruggeri <fruggeri@arista.com> Co-developed-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Dmitry Safonov <dima@arista.com> Acked-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* net/tcp: Add TCP-AO SNE supportDmitry Safonov2023-10-271-1/+2
| | | | | | | | | | | | | | Add Sequence Number Extension (SNE) for TCP-AO. This is needed to protect long-living TCP-AO connections from replaying attacks after sequence number roll-over, see RFC5925 (6.2). Co-developed-by: Francesco Ruggeri <fruggeri@arista.com> Signed-off-by: Francesco Ruggeri <fruggeri@arista.com> Co-developed-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Dmitry Safonov <dima@arista.com> Acked-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* net/tcp: Add TCP-AO segments countersDmitry Safonov2023-10-271-1/+1
| | | | | | | | | | | | | | Introduce segment counters that are useful for troubleshooting/debugging as well as for writing tests. Now there are global snmp counters as well as per-socket and per-key. Co-developed-by: Francesco Ruggeri <fruggeri@arista.com> Signed-off-by: Francesco Ruggeri <fruggeri@arista.com> Co-developed-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Dmitry Safonov <dima@arista.com> Acked-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* net/tcp: Verify inbound TCP-AO signed segmentsDmitry Safonov2023-10-271-5/+5
| | | | | | | | | | | | | | | | | | | | Now there is a common function to verify signature on TCP segments: tcp_inbound_hash(). It has checks for all possible cross-interactions with MD5 signs as well as with unsigned segments. The rules from RFC5925 are: (1) Any TCP segment can have at max only one signature. (2) TCP connections can't switch between using TCP-MD5 and TCP-AO. (3) TCP-AO connections can't stop using AO, as well as unsigned connections can't suddenly start using AO. Co-developed-by: Francesco Ruggeri <fruggeri@arista.com> Signed-off-by: Francesco Ruggeri <fruggeri@arista.com> Co-developed-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Dmitry Safonov <dima@arista.com> Acked-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* net/tcp: Sign SYN-ACK segments with TCP-AODmitry Safonov2023-10-271-0/+1
| | | | | | | | | | | | | | Similarly to RST segments, wire SYN-ACKs to TCP-AO. tcp_rsk_used_ao() is handy here to check if the request socket used AO and needs a signature on the outgoing segments. Co-developed-by: Francesco Ruggeri <fruggeri@arista.com> Signed-off-by: Francesco Ruggeri <fruggeri@arista.com> Co-developed-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Dmitry Safonov <dima@arista.com> Acked-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* net/tcp: Wire TCP-AO to request socketsDmitry Safonov2023-10-271-8/+58
| | | | | | | | | | | | | | | | | | Now when the new request socket is created from the listening socket, it's recorded what MKT was used by the peer. tcp_rsk_used_ao() is a new helper for checking if TCP-AO option was used to create the request socket. tcp_ao_copy_all_matching() will copy all keys that match the peer on the request socket, as well as preparing them for the usage (creating traffic keys). Co-developed-by: Francesco Ruggeri <fruggeri@arista.com> Signed-off-by: Francesco Ruggeri <fruggeri@arista.com> Co-developed-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Dmitry Safonov <dima@arista.com> Acked-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* net/tcp: Add TCP-AO sign to twskDmitry Safonov2023-10-271-19/+73
| | | | | | | | | | | | | | | | Add support for sockets in time-wait state. ao_info as well as all keys are inherited on transition to time-wait socket. The lifetime of ao_info is now protected by ref counter, so that tcp_ao_destroy_sock() will destruct it only when the last user is gone. Co-developed-by: Francesco Ruggeri <fruggeri@arista.com> Signed-off-by: Francesco Ruggeri <fruggeri@arista.com> Co-developed-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Dmitry Safonov <dima@arista.com> Acked-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* net/tcp: Add AO sign to RST packetsDmitry Safonov2023-10-271-13/+56
| | | | | | | | | | | | Wire up sending resets to TCP-AO hashing. Co-developed-by: Francesco Ruggeri <fruggeri@arista.com> Signed-off-by: Francesco Ruggeri <fruggeri@arista.com> Co-developed-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Dmitry Safonov <dima@arista.com> Acked-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* net/tcp: Add tcp_parse_auth_options()Dmitry Safonov2023-10-271-5/+10
| | | | | | | | | | | | | | | | | | | Introduce a helper that: (1) shares the common code with TCP-MD5 header options parsing (2) looks for hash signature only once for both TCP-MD5 and TCP-AO (3) fails with -EEXIST if any TCP sign option is present twice, see RFC5925 (2.2): ">> A single TCP segment MUST NOT have more than one TCP-AO in its options sequence. When multiple TCP-AOs appear, TCP MUST discard the segment." Co-developed-by: Francesco Ruggeri <fruggeri@arista.com> Signed-off-by: Francesco Ruggeri <fruggeri@arista.com> Co-developed-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Dmitry Safonov <dima@arista.com> Acked-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* net/tcp: Add TCP-AO sign to outgoing packetsDmitry Safonov2023-10-271-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Using precalculated traffic keys, sign TCP segments as prescribed by RFC5925. Per RFC, TCP header options are included in sign calculation: "The TCP header, by default including options, and where the TCP checksum and TCP-AO MAC fields are set to zero, all in network- byte order." (5.1.3) tcp_ao_hash_header() has exclude_options parameter to optionally exclude TCP header from hash calculation, as described in RFC5925 (9.1), this is needed for interaction with middleboxes that may change "some TCP options". This is wired up to AO key flags and setsockopt() later. Similarly to TCP-MD5 hash TCP segment fragments. From this moment a user can start sending TCP-AO signed segments with one of crypto ahash algorithms from supported by Linux kernel. It can have a user-specified MAC length, to either save TCP option header space or provide higher protection using a longer signature. The inbound segments are not yet verified, TCP-AO option is ignored and they are accepted. Co-developed-by: Francesco Ruggeri <fruggeri@arista.com> Signed-off-by: Francesco Ruggeri <fruggeri@arista.com> Co-developed-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Dmitry Safonov <dima@arista.com> Acked-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* net/tcp: Calculate TCP-AO traffic keysDmitry Safonov2023-10-271-0/+1
| | | | | | | | | | | | | | Add traffic key calculation the way it's described in RFC5926. Wire it up to tcp_finish_connect() and cache the new keys straight away on already established TCP connections. Co-developed-by: Francesco Ruggeri <fruggeri@arista.com> Signed-off-by: Francesco Ruggeri <fruggeri@arista.com> Co-developed-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Dmitry Safonov <dima@arista.com> Acked-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* net/tcp: Prevent TCP-MD5 with TCP-AO being setDmitry Safonov2023-10-271-3/+11
| | | | | | | | | | | | | | | | | | | | Be as conservative as possible: if there is TCP-MD5 key for a given peer regardless of L3 interface - don't allow setting TCP-AO key for the same peer. According to RFC5925, TCP-AO is supposed to replace TCP-MD5 and there can't be any switch between both on any connected tuple. Later it can be relaxed, if there's a use, but in the beginning restrict any intersection. Note: it's still should be possible to set both TCP-MD5 and TCP-AO keys on a listening socket for *different* peers. Co-developed-by: Francesco Ruggeri <fruggeri@arista.com> Signed-off-by: Francesco Ruggeri <fruggeri@arista.com> Co-developed-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Dmitry Safonov <dima@arista.com> Acked-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* net/tcp: Introduce TCP_AO setsockopt()sDmitry Safonov2023-10-271-2/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add 3 setsockopt()s: 1. TCP_AO_ADD_KEY to add a new Master Key Tuple (MKT) on a socket 2. TCP_AO_DEL_KEY to delete present MKT from a socket 3. TCP_AO_INFO to change flags, Current_key/RNext_key on a TCP-AO sk Userspace has to introduce keys on every socket it wants to use TCP-AO option on, similarly to TCP_MD5SIG/TCP_MD5SIG_EXT. RFC5925 prohibits definition of MKTs that would match the same peer, so do sanity checks on the data provided by userspace. Be as conservative as possible, including refusal of defining MKT on an established connection with no AO, removing the key in-use and etc. (1) and (2) are to be used by userspace key manager to add/remove keys. (3) main purpose is to set RNext_key, which (as prescribed by RFC5925) is the KeyID that will be requested in TCP-AO header from the peer to sign their segments with. At this moment the life of ao_info ends in tcp_v4_destroy_sock(). Co-developed-by: Francesco Ruggeri <fruggeri@arista.com> Signed-off-by: Francesco Ruggeri <fruggeri@arista.com> Co-developed-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Salam Noureddine <noureddine@arista.com> Signed-off-by: Dmitry Safonov <dima@arista.com> Acked-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* net/tcp: Prepare tcp_md5sig_pool for TCP-AODmitry Safonov2023-10-271-41/+56
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | TCP-AO, similarly to TCP-MD5, needs to allocate tfms on a slow-path, which is setsockopt() and use crypto ahash requests on fast paths, which are RX/TX softirqs. Also, it needs a temporary/scratch buffer for preparing the hash. Rework tcp_md5sig_pool in order to support other hashing algorithms than MD5. It will make it possible to share pre-allocated crypto_ahash descriptors and scratch area between all TCP hash users. Internally tcp_sigpool calls crypto_clone_ahash() API over pre-allocated crypto ahash tfm. Kudos to Herbert, who provided this new crypto API. I was a little concerned over GFP_ATOMIC allocations of ahash and crypto_request in RX/TX (see tcp_sigpool_start()), so I benchmarked both "backends" with different algorithms, using patched version of iperf3[2]. On my laptop with i7-7600U @ 2.80GHz: clone-tfm per-CPU-requests TCP-MD5 2.25 Gbits/sec 2.30 Gbits/sec TCP-AO(hmac(sha1)) 2.53 Gbits/sec 2.54 Gbits/sec TCP-AO(hmac(sha512)) 1.67 Gbits/sec 1.64 Gbits/sec TCP-AO(hmac(sha384)) 1.77 Gbits/sec 1.80 Gbits/sec TCP-AO(hmac(sha224)) 1.29 Gbits/sec 1.30 Gbits/sec TCP-AO(hmac(sha3-512)) 481 Mbits/sec 480 Mbits/sec TCP-AO(hmac(md5)) 2.07 Gbits/sec 2.12 Gbits/sec TCP-AO(hmac(rmd160)) 1.01 Gbits/sec 995 Mbits/sec TCP-AO(cmac(aes128)) [not supporetd yet] 2.11 Gbits/sec So, it seems that my concerns don't have strong grounds and per-CPU crypto_request allocation can be dropped/removed from tcp_sigpool once ciphers get crypto_clone_ahash() support. [1]: https://lore.kernel.org/all/ZDefxOq6Ax0JeTRH@gondor.apana.org.au/T/#u [2]: https://github.com/0x7f454c46/iperf/tree/tcp-md5-ao Signed-off-by: Dmitry Safonov <dima@arista.com> Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com> Acked-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* tcp: add support for usec resolution in TCP TS valuesEric Dumazet2023-10-231-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Back in 2015, Van Jacobson suggested to use usec resolution in TCP TS values. This has been implemented in our private kernels. Goals were : 1) better observability of delays in networking stacks. 2) better disambiguation of events based on TSval/ecr values. 3) building block for congestion control modules needing usec resolution. Back then we implemented a schem based on private SYN options to negotiate the feature. For upstream submission, we chose to use a route attribute, because this feature is probably going to be used in private networks [1] [2]. ip route add 10/8 ... features tcp_usec_ts Note that RFC 7323 recommends a "timestamp clock frequency in the range 1 ms to 1 sec per tick.", but also mentions "the maximum acceptable clock frequency is one tick every 59 ns." [1] Unfortunately RFC 7323 5.5 (Outdated Timestamps) suggests to invalidate TS.Recent values after a flow was idle for more than 24 days. This is the part making usec_ts a problem for peers following this recommendation for long living idle flows. [2] Attempts to standardize usec ts went nowhere: https://www.ietf.org/proceedings/97/slides/slides-97-tcpm-tcp-options-for-low-latency-00.pdf https://datatracker.ietf.org/doc/draft-wang-tcpm-low-latency-opt/ Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* tcp: replace tcp_time_stamp_raw()Eric Dumazet2023-10-231-2/+2
| | | | | | | | | | | | | | In preparation of usec TCP TS support, remove tcp_time_stamp_raw() in favor of tcp_clock_ts() helper. This helper will return a suitable 32bit result to feed TS values, depending on a socket field. Also add tcp_tw_tsval() and tcp_rsk_tsval() helpers to factorize the details. We do not yet support usec timestamps. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/netJakub Kicinski2023-10-191-0/+1
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Cross-merge networking fixes after downstream PR. net/mac80211/key.c 02e0e426a2fb ("wifi: mac80211: fix error path key leak") 2a8b665e6bcc ("wifi: mac80211: remove key_mtx") 7d6904bf26b9 ("Merge wireless into wireless-next") https://lore.kernel.org/all/20231012113648.46eea5ec@canb.auug.org.au/ Adjacent changes: drivers/net/ethernet/ti/Kconfig a602ee3176a8 ("net: ethernet: ti: Fix mixed module-builtin object") 98bdeae9502b ("net: cpmac: remove driver to prepare for platform removal") Signed-off-by: Jakub Kicinski <kuba@kernel.org>
| * tcp: check mptcp-level constraints for backlog coalescingPaolo Abeni2023-10-191-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The MPTCP protocol can acquire the subflow-level socket lock and cause the tcp backlog usage. When inserting new skbs into the backlog, the stack will try to coalesce them. Currently, we have no check in place to ensure that such coalescing will respect the MPTCP-level DSS, and that may cause data stream corruption, as reported by Christoph. Address the issue by adding the relevant admission check for coalescing in tcp_add_backlog(). Note the issue is not easy to reproduce, as the MPTCP protocol tries hard to avoid acquiring the subflow-level socket lock. Fixes: 648ef4b88673 ("mptcp: Implement MPTCP receive path") Cc: stable@vger.kernel.org Reported-by: Christoph Paasch <cpaasch@apple.com> Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/420 Reviewed-by: Mat Martineau <martineau@kernel.org> Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: Mat Martineau <martineau@kernel.org> Link: https://lore.kernel.org/r/20231018-send-net-20231018-v1-2-17ecb002e41d@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
* | Merge tag 'for-netdev' of ↵Jakub Kicinski2023-10-161-1/+1
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next Daniel Borkmann says: ==================== pull-request: bpf-next 2023-10-16 We've added 90 non-merge commits during the last 25 day(s) which contain a total of 120 files changed, 3519 insertions(+), 895 deletions(-). The main changes are: 1) Add missed stats for kprobes to retrieve the number of missed kprobe executions and subsequent executions of BPF programs, from Jiri Olsa. 2) Add cgroup BPF sockaddr hooks for unix sockets. The use case is for systemd to reimplement the LogNamespace feature which allows running multiple instances of systemd-journald to process the logs of different services, from Daan De Meyer. 3) Implement BPF CPUv4 support for s390x BPF JIT, from Ilya Leoshkevich. 4) Improve BPF verifier log output for scalar registers to better disambiguate their internal state wrt defaults vs min/max values matching, from Andrii Nakryiko. 5) Extend the BPF fib lookup helpers for IPv4/IPv6 to support retrieving the source IP address with a new BPF_FIB_LOOKUP_SRC flag, from Martynas Pumputis. 6) Add support for open-coded task_vma iterator to help with symbolization for BPF-collected user stacks, from Dave Marchevsky. 7) Add libbpf getters for accessing individual BPF ring buffers which is useful for polling them individually, for example, from Martin Kelly. 8) Extend AF_XDP selftests to validate the SHARED_UMEM feature, from Tushar Vyavahare. 9) Improve BPF selftests cross-building support for riscv arch, from Björn Töpel. 10) Add the ability to pin a BPF timer to the same calling CPU, from David Vernet. 11) Fix libbpf's bpf_tracing.h macros for riscv to use the generic implementation of PT_REGS_SYSCALL_REGS() to access syscall arguments, from Alexandre Ghiti. 12) Extend libbpf to support symbol versioning for uprobes, from Hengqi Chen. 13) Fix bpftool's skeleton code generation to guarantee that ELF data is 8 byte aligned, from Ian Rogers. 14) Inherit system-wide cpu_mitigations_off() setting for Spectre v1/v4 security mitigations in BPF verifier, from Yafang Shao. 15) Annotate struct bpf_stack_map with __counted_by attribute to prepare BPF side for upcoming __counted_by compiler support, from Kees Cook. * tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next: (90 commits) bpf: Ensure proper register state printing for cond jumps bpf: Disambiguate SCALAR register state output in verifier logs selftests/bpf: Make align selftests more robust selftests/bpf: Improve missed_kprobe_recursion test robustness selftests/bpf: Improve percpu_alloc test robustness selftests/bpf: Add tests for open-coded task_vma iter bpf: Introduce task_vma open-coded iterator kfuncs selftests/bpf: Rename bpf_iter_task_vma.c to bpf_iter_task_vmas.c bpf: Don't explicitly emit BTF for struct btf_iter_num bpf: Change syscall_nr type to int in struct syscall_tp_t net/bpf: Avoid unused "sin_addr_len" warning when CONFIG_CGROUP_BPF is not set bpf: Avoid unnecessary audit log for CPU security mitigations selftests/bpf: Add tests for cgroup unix socket address hooks selftests/bpf: Make sure mount directory exists documentation/bpf: Document cgroup unix socket address hooks bpftool: Add support for cgroup unix socket address hooks libbpf: Add support for cgroup unix socket address hooks bpf: Implement cgroup sockaddr hooks for unix sockets bpf: Add bpf_sock_addr_set_sun_path() to allow writing unix sockaddr from bpf bpf: Propagate modified uaddrlen from cgroup sockaddr programs ... ==================== Link: https://lore.kernel.org/r/20231016204803.30153-1-daniel@iogearbox.net Signed-off-by: Jakub Kicinski <kuba@kernel.org>
| * | bpf: Propagate modified uaddrlen from cgroup sockaddr programsDaan De Meyer2023-10-111-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | As prep for adding unix socket support to the cgroup sockaddr hooks, let's propagate the sockaddr length back to the caller after running a bpf cgroup sockaddr hook program. While not important for AF_INET or AF_INET6, the sockaddr length is important when working with AF_UNIX sockaddrs as the size of the sockaddr cannot be determined just from the address family or the sockaddr's contents. __cgroup_bpf_run_filter_sock_addr() is modified to take the uaddrlen as an input/output argument. After running the program, the modified sockaddr length is stored in the uaddrlen pointer. Signed-off-by: Daan De Meyer <daan.j.demeyer@gmail.com> Link: https://lore.kernel.org/r/20231011185113.140426-3-daan.j.demeyer@gmail.com Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
* | | tcp: Set pingpong threshold via sysctlHaiyang Zhang2023-10-161-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | TCP pingpong threshold is 1 by default. But some applications, like SQL DB may prefer a higher pingpong threshold to activate delayed acks in quick ack mode for better performance. The pingpong threshold and related code were changed to 3 in the year 2019 in: commit 4a41f453bedf ("tcp: change pingpong threshold to 3") And reverted to 1 in the year 2022 in: commit 4d8f24eeedc5 ("Revert "tcp: change pingpong threshold to 3"") There is no single value that fits all applications. Add net.ipv4.tcp_pingpong_thresh sysctl tunable, so it can be tuned for optimal performance based on the application needs. Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com> Reviewed-by: Simon Horman <horms@kernel.org> Reviewed-by: Eric Dumazet <edumazet@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com> Link: https://lore.kernel.org/r/1697056244-21888-1-git-send-email-haiyangz@microsoft.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
* | | inet: implement lockless IP_TOSEric Dumazet2023-10-011-4/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Some reads of inet->tos are racy. Add needed READ_ONCE() annotations and convert IP_TOS option lockless. v2: missing changes in include/net/route.h (David Ahern) Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* | | net: implement lockless SO_PRIORITYEric Dumazet2023-10-011-1/+1
|/ / | | | | | | | | | | | | | | | | | | | | This is a followup of 8bf43be799d4 ("net: annotate data-races around sk->sk_priority"). sk->sk_priority can be read and written without holding the socket lock. Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Wenjia Zhang <wenjia@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* / tcp: defer regular ACK while processing socket backlogEric Dumazet2023-09-121-0/+1
|/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This idea came after a particular workload requested the quickack attribute set on routes, and a performance drop was noticed for large bulk transfers. For high throughput flows, it is best to use one cpu running the user thread issuing socket system calls, and a separate cpu to process incoming packets from BH context. (With TSO/GRO, bottleneck is usually the 'user' cpu) Problem is the user thread can spend a lot of time while holding the socket lock, forcing BH handler to queue most of incoming packets in the socket backlog. Whenever the user thread releases the socket lock, it must first process all accumulated packets in the backlog, potentially adding latency spikes. Due to flood mitigation, having too many packets in the backlog increases chance of unexpected drops. Backlog processing unfortunately shifts a fair amount of cpu cycles from the BH cpu to the 'user' cpu, thus reducing max throughput. This patch takes advantage of the backlog processing, and the fact that ACK are mostly cumulative. The idea is to detect we are in the backlog processing and defer all eligible ACK into a single one, sent from tcp_release_cb(). This saves cpu cycles on both sides, and network resources. Performance of a single TCP flow on a 200Gbit NIC: - Throughput is increased by 20% (100Gbit -> 120Gbit). - Number of generated ACK per second shrinks from 240,000 to 40,000. - Number of backlog drops per second shrinks from 230 to 0. Benchmark context: - Regular netperf TCP_STREAM (no zerocopy) - Intel(R) Xeon(R) Platinum 8481C (Saphire Rapids) - MAX_SKB_FRAGS = 17 (~60KB per GRO packet) This feature is guarded by a new sysctl, and enabled by default: /proc/sys/net/ipv4/tcp_backlog_ack_defer Signed-off-by: Eric Dumazet <edumazet@google.com> Acked-by: Yuchung Cheng <ycheng@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Acked-by: Soheil Hassas Yeganeh <soheil@google.com> Acked-by: Dave Taht <dave.taht@gmail.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
* Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/netJakub Kicinski2023-08-241-2/+2
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Cross-merge networking fixes after downstream PR. Conflicts: include/net/inet_sock.h f866fbc842de ("ipv4: fix data-races around inet->inet_id") c274af224269 ("inet: introduce inet->inet_flags") https://lore.kernel.org/all/679ddff6-db6e-4ff6-b177-574e90d0103d@tessares.net/ Adjacent changes: drivers/net/bonding/bond_alb.c e74216b8def3 ("bonding: fix macvlan over alb bond support") f11e5bd159b0 ("bonding: support balance-alb with openvswitch") drivers/net/ethernet/broadcom/bgmac.c d6499f0b7c7c ("net: bgmac: Return PTR_ERR() for fixed_phy_register()") 23a14488ea58 ("net: bgmac: Fix return value check for fixed_phy_register()") drivers/net/ethernet/broadcom/genet/bcmmii.c 32bbe64a1386 ("net: bcmgenet: Fix return value check for fixed_phy_register()") acf50d1adbf4 ("net: bcmgenet: Return PTR_ERR() for fixed_phy_register()") net/sctp/socket.c f866fbc842de ("ipv4: fix data-races around inet->inet_id") b09bde5c3554 ("inet: move inet->mc_loop to inet->inet_frags") Signed-off-by: Jakub Kicinski <kuba@kernel.org>
| * ipv4: fix data-races around inet->inet_idEric Dumazet2023-08-201-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | UDP sendmsg() is lockless, so ip_select_ident_segs() can very well be run from multiple cpus [1] Convert inet->inet_id to an atomic_t, but implement a dedicated path for TCP, avoiding cost of a locked instruction (atomic_add_return()) Note that this patch will cause a trivial merge conflict because we added inet->flags in net-next tree. v2: added missing change in drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c (David Ahern) [1] BUG: KCSAN: data-race in __ip_make_skb / __ip_make_skb read-write to 0xffff888145af952a of 2 bytes by task 7803 on cpu 1: ip_select_ident_segs include/net/ip.h:542 [inline] ip_select_ident include/net/ip.h:556 [inline] __ip_make_skb+0x844/0xc70 net/ipv4/ip_output.c:1446 ip_make_skb+0x233/0x2c0 net/ipv4/ip_output.c:1560 udp_sendmsg+0x1199/0x1250 net/ipv4/udp.c:1260 inet_sendmsg+0x63/0x80 net/ipv4/af_inet.c:830 sock_sendmsg_nosec net/socket.c:725 [inline] sock_sendmsg net/socket.c:748 [inline] ____sys_sendmsg+0x37c/0x4d0 net/socket.c:2494 ___sys_sendmsg net/socket.c:2548 [inline] __sys_sendmmsg+0x269/0x500 net/socket.c:2634 __do_sys_sendmmsg net/socket.c:2663 [inline] __se_sys_sendmmsg net/socket.c:2660 [inline] __x64_sys_sendmmsg+0x57/0x60 net/socket.c:2660 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd read to 0xffff888145af952a of 2 bytes by task 7804 on cpu 0: ip_select_ident_segs include/net/ip.h:541 [inline] ip_select_ident include/net/ip.h:556 [inline] __ip_make_skb+0x817/0xc70 net/ipv4/ip_output.c:1446 ip_make_skb+0x233/0x2c0 net/ipv4/ip_output.c:1560 udp_sendmsg+0x1199/0x1250 net/ipv4/udp.c:1260 inet_sendmsg+0x63/0x80 net/ipv4/af_inet.c:830 sock_sendmsg_nosec net/socket.c:725 [inline] sock_sendmsg net/socket.c:748 [inline] ____sys_sendmsg+0x37c/0x4d0 net/socket.c:2494 ___sys_sendmsg net/socket.c:2548 [inline] __sys_sendmmsg+0x269/0x500 net/socket.c:2634 __do_sys_sendmmsg net/socket.c:2663 [inline] __se_sys_sendmmsg net/socket.c:2660 [inline] __x64_sys_sendmmsg+0x57/0x60 net/socket.c:2660 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd value changed: 0x184d -> 0x184e Reported by Kernel Concurrency Sanitizer on: CPU: 0 PID: 7804 Comm: syz-executor.1 Not tainted 6.5.0-rc6-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/26/2023 ================================================================== Fixes: 23f57406b82d ("ipv4: avoid using shared IP generator for connected sockets") Reported-by: syzbot <syzkaller@googlegroups.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: David Ahern <dsahern@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* | inet: move inet->recverr to inet->inet_flagsEric Dumazet2023-08-161-3/+2
| | | | | | | | | | | | | | | | | | | | | | IP_RECVERR socket option can now be set/get without locking the socket. This patch potentially avoid data-races around inet->recverr. Signed-off-by: Eric Dumazet <edumazet@google.com> Acked-by: Soheil Hassas Yeganeh <soheil@google.com> Reviewed-by: Simon Horman <horms@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* | Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/netJakub Kicinski2023-08-031-2/+2
|\| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Cross-merge networking fixes after downstream PR. Conflicts: net/dsa/port.c 9945c1fb03a3 ("net: dsa: fix older DSA drivers using phylink") a88dd7538461 ("net: dsa: remove legacy_pre_march2020 detection") https://lore.kernel.org/all/20230731102254.2c9868ca@canb.auug.org.au/ net/xdp/xsk.c 3c5b4d69c358 ("net: annotate data-races around sk->sk_mark") b7f72a30e9ac ("xsk: introduce wrappers and helpers for supporting multi-buffer in Tx path") https://lore.kernel.org/all/20230731102631.39988412@canb.auug.org.au/ drivers/net/ethernet/broadcom/bnxt/bnxt.c 37b61cda9c16 ("bnxt: don't handle XDP in netpoll") 2b56b3d99241 ("eth: bnxt: handle invalid Tx completions more gracefully") https://lore.kernel.org/all/20230801101708.1dc7faac@canb.auug.org.au/ Adjacent changes: drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_fs.c 62da08331f1a ("net/mlx5e: Set proper IPsec source port in L4 selector") fbd517549c32 ("net/mlx5e: Add function to get IPsec offload namespace") drivers/net/ethernet/sfc/selftest.c 55c1528f9b97 ("sfc: fix field-spanning memcpy in selftest") ae9d445cd41f ("sfc: Miscellaneous comment removals") Signed-off-by: Jakub Kicinski <kuba@kernel.org>
| * net: annotate data-races around sk->sk_priorityEric Dumazet2023-07-291-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | sk_getsockopt() runs locklessly. This means sk->sk_priority can be read while other threads are changing its value. Other reads also happen without socket lock being held. Add missing annotations where needed. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-19 14:29:13 +0000