From 2397849baa7c44c242e5d5142d5d16d1e7ed53d0 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sat, 9 Jun 2012 14:56:12 -0700 Subject: [PATCH] tcp: Cache inetpeer in timewait socket, and only when necessary. Since it's guarenteed that we will access the inetpeer if we're trying to do timewait recycling and TCP options were enabled on the connection, just cache the peer in the timewait socket. In the future, inetpeer lookups will be context dependent (per routing realm), and this helps facilitate that as well. Signed-off-by: David S. Miller --- include/linux/tcp.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux/tcp.h') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 4c5b63283377..23e8234f75a5 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -506,8 +506,9 @@ struct tcp_timewait_sock { u32 tw_rcv_wnd; u32 tw_ts_recent; long tw_ts_recent_stamp; + struct inet_peer *tw_peer; #ifdef CONFIG_TCP_MD5SIG - struct tcp_md5sig_key *tw_md5_key; + struct tcp_md5sig_key *tw_md5_key; #endif /* Few sockets in timewait have cookies; in that case, then this * object holds a reference to them (tw_cookie_values->kref). -- cgit From b6242b9b45e84ef71c59002cd128c3197938cb2f Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 10 Jul 2012 03:27:56 -0700 Subject: tcp: Remove tw->tw_peer No longer used. Signed-off-by: David S. Miller --- include/linux/tcp.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux/tcp.h') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 7d3bcedc062a..2de9cf46f9fc 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -506,7 +506,6 @@ struct tcp_timewait_sock { u32 tw_rcv_wnd; u32 tw_ts_recent; long tw_ts_recent_stamp; - struct inet_peer *tw_peer; #ifdef CONFIG_TCP_MD5SIG struct tcp_md5sig_key *tw_md5_key; #endif -- cgit From 46d3ceabd8d98ed0ad10f20c595ca784e34786c5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 11 Jul 2012 05:50:31 +0000 Subject: tcp: TCP Small Queues This introduce TSQ (TCP Small Queues) TSQ goal is to reduce number of TCP packets in xmit queues (qdisc & device queues), to reduce RTT and cwnd bias, part of the bufferbloat problem. sk->sk_wmem_alloc not allowed to grow above a given limit, allowing no more than ~128KB [1] per tcp socket in qdisc/dev layers at a given time. TSO packets are sized/capped to half the limit, so that we have two TSO packets in flight, allowing better bandwidth use. As a side effect, setting the limit to 40000 automatically reduces the standard gso max limit (65536) to 40000/2 : It can help to reduce latencies of high prio packets, having smaller TSO packets. This means we divert sock_wfree() to a tcp_wfree() handler, to queue/send following frames when skb_orphan() [2] is called for the already queued skbs. Results on my dev machines (tg3/ixgbe nics) are really impressive, using standard pfifo_fast, and with or without TSO/GSO. Without reduction of nominal bandwidth, we have reduction of buffering per bulk sender : < 1ms on Gbit (instead of 50ms with TSO) < 8ms on 100Mbit (instead of 132 ms) I no longer have 4 MBytes backlogged in qdisc by a single netperf session, and both side socket autotuning no longer use 4 Mbytes. As skb destructor cannot restart xmit itself ( as qdisc lock might be taken at this point ), we delegate the work to a tasklet. We use one tasklest per cpu for performance reasons. If tasklet finds a socket owned by the user, it sets TSQ_OWNED flag. This flag is tested in a new protocol method called from release_sock(), to eventually send new segments. [1] New /proc/sys/net/ipv4/tcp_limit_output_bytes tunable [2] skb_orphan() is usually called at TX completion time, but some drivers call it in their start_xmit() handler. These drivers should at least use BQL, or else a single TCP session can still fill the whole NIC TX ring, since TSQ will have no effect. Signed-off-by: Eric Dumazet Cc: Dave Taht Cc: Tom Herbert Cc: Matt Mathis Cc: Yuchung Cheng Cc: Nandita Dukkipati Signed-off-by: David S. Miller --- include/linux/tcp.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/linux/tcp.h') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 2de9cf46f9fc..1888169e07c7 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -339,6 +339,9 @@ struct tcp_sock { u32 rcv_tstamp; /* timestamp of last received ACK (for keepalives) */ u32 lsndtime; /* timestamp of last sent data packet (for restart window) */ + struct list_head tsq_node; /* anchor in tsq_tasklet.head list */ + unsigned long tsq_flags; + /* Data for direct copy to user */ struct { struct sk_buff_head prequeue; @@ -494,6 +497,12 @@ struct tcp_sock { struct tcp_cookie_values *cookie_values; }; +enum tsq_flags { + TSQ_THROTTLED, + TSQ_QUEUED, + TSQ_OWNED, /* tcp_tasklet_func() found socket was locked */ +}; + static inline struct tcp_sock *tcp_sk(const struct sock *sk) { return (struct tcp_sock *)sk; -- cgit From 2100c8d2d9db23c0a09901a782bb4e3b21bee298 Mon Sep 17 00:00:00 2001 From: Yuchung Cheng Date: Thu, 19 Jul 2012 06:43:05 +0000 Subject: net-tcp: Fast Open base This patch impelements the common code for both the client and server. 1. TCP Fast Open option processing. Since Fast Open does not have an option number assigned by IANA yet, it shares the experiment option code 254 by implementing draft-ietf-tcpm-experimental-options with a 16 bits magic number 0xF989. This enables global experiments without clashing the scarce(2) experimental options available for TCP. When the draft status becomes standard (maybe), the client should switch to the new option number assigned while the server supports both numbers for transistion. 2. The new sysctl tcp_fastopen 3. A place holder init function Signed-off-by: Yuchung Cheng Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/tcp.h | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'include/linux/tcp.h') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 1888169e07c7..12948f543839 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -243,6 +243,16 @@ static inline unsigned int tcp_optlen(const struct sk_buff *skb) return (tcp_hdr(skb)->doff - 5) * 4; } +/* TCP Fast Open */ +#define TCP_FASTOPEN_COOKIE_MIN 4 /* Min Fast Open Cookie size in bytes */ +#define TCP_FASTOPEN_COOKIE_MAX 16 /* Max Fast Open Cookie size in bytes */ + +/* TCP Fast Open Cookie as stored in memory */ +struct tcp_fastopen_cookie { + s8 len; + u8 val[TCP_FASTOPEN_COOKIE_MAX]; +}; + /* This defines a selective acknowledgement block. */ struct tcp_sack_block_wire { __be32 start_seq; -- cgit From 783237e8daf13481ee234997cbbbb823872ac388 Mon Sep 17 00:00:00 2001 From: Yuchung Cheng Date: Thu, 19 Jul 2012 06:43:07 +0000 Subject: net-tcp: Fast Open client - sending SYN-data This patch implements sending SYN-data in tcp_connect(). The data is from tcp_sendmsg() with flag MSG_FASTOPEN (implemented in a later patch). The length of the cookie in tcp_fastopen_req, init'd to 0, controls the type of the SYN. If the cookie is not cached (len==0), the host sends data-less SYN with Fast Open cookie request option to solicit a cookie from the remote. If cookie is not available (len > 0), the host sends a SYN-data with Fast Open cookie option. If cookie length is negative, the SYN will not include any Fast Open option (for fall back operations). To deal with middleboxes that may drop SYN with data or experimental TCP option, the SYN-data is only sent once. SYN retransmits do not include data or Fast Open options. The connection will fall back to regular TCP handshake. Signed-off-by: Yuchung Cheng Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/tcp.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'include/linux/tcp.h') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 12948f543839..1edf96afab44 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -386,7 +386,8 @@ struct tcp_sock { unused : 1; u8 repair_queue; u8 do_early_retrans:1,/* Enable RFC5827 early-retransmit */ - early_retrans_delayed:1; /* Delayed ER timer installed */ + early_retrans_delayed:1, /* Delayed ER timer installed */ + syn_fastopen:1; /* SYN includes Fast Open option */ /* RTT measurement */ u32 srtt; /* smoothed round trip time << 3 */ @@ -500,6 +501,9 @@ struct tcp_sock { struct tcp_md5sig_info __rcu *md5sig_info; #endif +/* TCP fastopen related information */ + struct tcp_fastopen_request *fastopen_req; + /* When the cookie options are generated and exchanged, then this * object holds a reference to them (cookie_values->kref). Also * contains related tcp_cookie_transactions fields. -- cgit From 67da22d23fa6f3324e03bcd0580b914b2e4afbf3 Mon Sep 17 00:00:00 2001 From: Yuchung Cheng Date: Thu, 19 Jul 2012 06:43:11 +0000 Subject: net-tcp: Fast Open client - cookie-less mode In trusted networks, e.g., intranet, data-center, the client does not need to use Fast Open cookie to mitigate DoS attacks. In cookie-less mode, sendmsg() with MSG_FASTOPEN flag will send SYN-data regardless of cookie availability. Signed-off-by: Yuchung Cheng Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/tcp.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux/tcp.h') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 1edf96afab44..9febfb685c33 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -387,6 +387,7 @@ struct tcp_sock { u8 repair_queue; u8 do_early_retrans:1,/* Enable RFC5827 early-retransmit */ early_retrans_delayed:1, /* Delayed ER timer installed */ + syn_data:1, /* SYN includes data */ syn_fastopen:1; /* SYN includes Fast Open option */ /* RTT measurement */ -- cgit From 6f458dfb409272082c9bfa412f77ff2fc21c626f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 20 Jul 2012 05:45:50 +0000 Subject: tcp: improve latencies of timer triggered events Modern TCP stack highly depends on tcp_write_timer() having a small latency, but current implementation doesn't exactly meet the expectations. When a timer fires but finds the socket is owned by the user, it rearms itself for an additional delay hoping next run will be more successful. tcp_write_timer() for example uses a 50ms delay for next try, and it defeats many attempts to get predictable TCP behavior in term of latencies. Use the recently introduced tcp_release_cb(), so that the user owning the socket will call various handlers right before socket release. This will permit us to post a followup patch to address the tcp_tso_should_defer() syndrome (some deferred packets have to wait RTO timer to be transmitted, while cwnd should allow us to send them sooner) Signed-off-by: Eric Dumazet Cc: Tom Herbert Cc: Yuchung Cheng Cc: Neal Cardwell Cc: Nandita Dukkipati Cc: H.K. Jerry Chu Cc: John Heffner Signed-off-by: David S. Miller --- include/linux/tcp.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include/linux/tcp.h') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 9febfb685c33..2761856987b2 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -515,7 +515,9 @@ struct tcp_sock { enum tsq_flags { TSQ_THROTTLED, TSQ_QUEUED, - TSQ_OWNED, /* tcp_tasklet_func() found socket was locked */ + TCP_TSQ_DEFERRED, /* tcp_tasklet_func() found socket was owned */ + TCP_WRITE_TIMER_DEFERRED, /* tcp_write_timer() found socket was owned */ + TCP_DELACK_TIMER_DEFERRED, /* tcp_delack_timer() found socket was owned */ }; static inline struct tcp_sock *tcp_sk(const struct sock *sk) -- cgit From 563d34d05786263893ba4a1042eb9b9374127cf5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 23 Jul 2012 09:48:52 +0200 Subject: tcp: dont drop MTU reduction indications MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit ICMP messages generated in output path if frame length is bigger than mtu are actually lost because socket is owned by user (doing the xmit) One example is the ipgre_tunnel_xmit() calling icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu)); We had a similar case fixed in commit a34a101e1e6 (ipv6: disable GSO on sockets hitting dst_allfrag). Problem of such fix is that it relied on retransmit timers, so short tcp sessions paid a too big latency increase price. This patch uses the tcp_release_cb() infrastructure so that MTU reduction messages (ICMP messages) are not lost, and no extra delay is added in TCP transmits. Reported-by: Maciej Żenczykowski Diagnosed-by: Neal Cardwell Signed-off-by: Eric Dumazet Cc: Nandita Dukkipati Cc: Tom Herbert Cc: Tore Anderson Signed-off-by: David S. Miller --- include/linux/tcp.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include/linux/tcp.h') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 2761856987b2..eb125a4c30b3 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -493,6 +493,9 @@ struct tcp_sock { u32 probe_seq_start; u32 probe_seq_end; } mtu_probe; + u32 mtu_info; /* We received an ICMP_FRAG_NEEDED / ICMPV6_PKT_TOOBIG + * while socket was owned by user. + */ #ifdef CONFIG_TCP_MD5SIG /* TCP AF-Specific parts; only used by MD5 Signature support so far */ @@ -518,6 +521,9 @@ enum tsq_flags { TCP_TSQ_DEFERRED, /* tcp_tasklet_func() found socket was owned */ TCP_WRITE_TIMER_DEFERRED, /* tcp_write_timer() found socket was owned */ TCP_DELACK_TIMER_DEFERRED, /* tcp_delack_timer() found socket was owned */ + TCP_MTU_REDUCED_DEFERRED, /* tcp_v{4|6}_err() could not call + * tcp_v{4|6}_mtu_reduced() + */ }; static inline struct tcp_sock *tcp_sk(const struct sock *sk) -- cgit