From 492135557dc090a1abb2cfbe1a412757e3ed68ab Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Tue, 19 May 2015 21:04:22 +0200 Subject: [PATCH] tcp: add rfc3168, section 6.1.1.1. fallback MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit This work as a follow-up of commit f7b3bec6f516 ("net: allow setting ecn via routing table") and adds RFC3168 section 6.1.1.1. fallback for outgoing ECN connections. In other words, this work adds a retry with a non-ECN setup SYN packet, as suggested from the RFC on the first timeout: [...] A host that receives no reply to an ECN-setup SYN within the normal SYN retransmission timeout interval MAY resend the SYN and any subsequent SYN retransmissions with CWR and ECE cleared. [...] Schematic client-side view when assuming the server is in tcp_ecn=2 mode, that is, Linux default since 2009 via commit 255cac91c3c9 ("tcp: extend ECN sysctl to allow server-side only ECN"): 1) Normal ECN-capable path: SYN ECE CWR -----> <----- SYN ACK ECE ACK -----> 2) Path with broken middlebox, when client has fallback: SYN ECE CWR ----X crappy middlebox drops packet (timeout, rtx) SYN -----> <----- SYN ACK ACK -----> In case we would not have the fallback implemented, the middlebox drop point would basically end up as: SYN ECE CWR ----X crappy middlebox drops packet (timeout, rtx) SYN ECE CWR ----X crappy middlebox drops packet (timeout, rtx) SYN ECE CWR ----X crappy middlebox drops packet (timeout, rtx) In any case, it's rather a smaller percentage of sites where there would occur such additional setup latency: it was found in end of 2014 that ~56% of IPv4 and 65% of IPv6 servers of Alexa 1 million list would negotiate ECN (aka tcp_ecn=2 default), 0.42% of these webservers will fail to connect when trying to negotiate with ECN (tcp_ecn=1) due to timeouts, which the fallback would mitigate with a slight latency trade-off. Recent related paper on this topic: Brian Trammell, Mirja Kühlewind, Damiano Boppart, Iain Learmonth, Gorry Fairhurst, and Richard Scheffenegger: "Enabling Internet-Wide Deployment of Explicit Congestion Notification." Proc. PAM 2015, New York. http://ecn.ethz.ch/ecn-pam15.pdf Thus, when net.ipv4.tcp_ecn=1 is being set, the patch will perform RFC3168, section 6.1.1.1. fallback on timeout. For users explicitly not wanting this which can be in DC use case, we add a net.ipv4.tcp_ecn_fallback knob that allows for disabling the fallback. tp->ecn_flags are not being cleared in tcp_ecn_clear_syn() on output, but rather we let tcp_ecn_rcv_synack() take that over on input path in case a SYN ACK ECE was delayed. Thus a spurious SYN retransmission will not prevent ECN being negotiated eventually in that case. Reference: https://www.ietf.org/proceedings/92/slides/slides-92-iccrg-1.pdf Reference: https://www.ietf.org/proceedings/89/slides/slides-89-tsvarea-1.pdf Signed-off-by: Daniel Borkmann Signed-off-by: Florian Westphal Signed-off-by: Mirja Kühlewind Signed-off-by: Brian Trammell Cc: Eric Dumazet Cc: Dave That Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- Documentation/networking/dctcp.txt | 1 + Documentation/networking/ip-sysctl.txt | 9 +++++++++ include/net/netns/ipv4.h | 2 ++ include/net/tcp.h | 2 ++ net/ipv4/sysctl_net_ipv4.c | 7 +++++++ net/ipv4/tcp_ipv4.c | 5 ++++- net/ipv4/tcp_output.c | 13 +++++++++++++ 7 files changed, 38 insertions(+), 1 deletion(-) diff --git a/Documentation/networking/dctcp.txt b/Documentation/networking/dctcp.txt index 0d5dfbc89ec9..13a857753208 100644 --- a/Documentation/networking/dctcp.txt +++ b/Documentation/networking/dctcp.txt @@ -8,6 +8,7 @@ the data center network to provide multi-bit feedback to the end hosts. To enable it on end hosts: sysctl -w net.ipv4.tcp_congestion_control=dctcp + sysctl -w net.ipv4.tcp_ecn_fallback=0 (optional) All switches in the data center network running DCTCP must support ECN marking and be configured for marking when reaching defined switch buffer diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index 5095c63e50ed..cb083e0d682c 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -267,6 +267,15 @@ tcp_ecn - INTEGER but do not request ECN on outgoing connections. Default: 2 +tcp_ecn_fallback - BOOLEAN + If the kernel detects that ECN connection misbehaves, enable fall + back to non-ECN. Currently, this knob implements the fallback + from RFC3168, section 6.1.1.1., but we reserve that in future, + additional detection mechanisms could be implemented under this + knob. The value is not used, if tcp_ecn or per route (or congestion + control) ECN settings are disabled. + Default: 1 (fallback enabled) + tcp_fack - BOOLEAN Enable FACK congestion avoidance and fast retransmission. The value is not used, if tcp_sack is not enabled. diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 614a49be68a9..6848b8bb2e63 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -77,6 +77,8 @@ struct netns_ipv4 { struct local_ports ip_local_ports; int sysctl_tcp_ecn; + int sysctl_tcp_ecn_fallback; + int sysctl_ip_no_pmtu_disc; int sysctl_ip_fwd_use_pmtu; int sysctl_ip_nonlocal_bind; diff --git a/include/net/tcp.h b/include/net/tcp.h index 0d85223efa4c..2bb2bad21d5c 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -712,6 +712,8 @@ static inline u32 tcp_skb_timestamp(const struct sk_buff *skb) #define TCPHDR_ECE 0x40 #define TCPHDR_CWR 0x80 +#define TCPHDR_SYN_ECN (TCPHDR_SYN | TCPHDR_ECE | TCPHDR_CWR) + /* This is what the send packet queuing engine uses to pass * TCP per-packet control information to the transmission code. * We also store the host-order sequence numbers in here too. diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index c3852a7ff3c7..841de32f1fee 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -820,6 +820,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_ecn_fallback", + .data = &init_net.ipv4.sysctl_tcp_ecn_fallback, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { .procname = "ip_local_port_range", .maxlen = sizeof(init_net.ipv4.ip_local_ports.range), diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 91cb4768a860..0cc4b5a630cd 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2411,12 +2411,15 @@ static int __net_init tcp_sk_init(struct net *net) goto fail; *per_cpu_ptr(net->ipv4.tcp_sk, cpu) = sk; } + net->ipv4.sysctl_tcp_ecn = 2; + net->ipv4.sysctl_tcp_ecn_fallback = 1; + net->ipv4.sysctl_tcp_base_mss = TCP_BASE_MSS; net->ipv4.sysctl_tcp_probe_threshold = TCP_PROBE_THRESHOLD; net->ipv4.sysctl_tcp_probe_interval = TCP_PROBE_INTERVAL; - return 0; + return 0; fail: tcp_sk_exit(net); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 08c2cc40b26d..e29d43b5a0bb 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -350,6 +350,15 @@ static void tcp_ecn_send_syn(struct sock *sk, struct sk_buff *skb) } } +static void tcp_ecn_clear_syn(struct sock *sk, struct sk_buff *skb) +{ + if (sock_net(sk)->ipv4.sysctl_tcp_ecn_fallback) + /* tp->ecn_flags are cleared at a later point in time when + * SYN ACK is ultimatively being received. + */ + TCP_SKB_CB(skb)->tcp_flags &= ~(TCPHDR_ECE | TCPHDR_CWR); +} + static void tcp_ecn_make_synack(const struct request_sock *req, struct tcphdr *th, struct sock *sk) @@ -2615,6 +2624,10 @@ int __tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb) } } + /* RFC3168, section 6.1.1.1. ECN fallback */ + if ((TCP_SKB_CB(skb)->tcp_flags & TCPHDR_SYN_ECN) == TCPHDR_SYN_ECN) + tcp_ecn_clear_syn(sk, skb); + tcp_retrans_try_collapse(sk, skb, cur_mss); /* Make a copy, if the first transmission SKB clone we made -- 2.20.1