From 74abc20cedf6f95a0065665c55180ef2d4828ba5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Feb 2015 19:08:59 -0800 Subject: tcp: cleanup static functions tcp_fastopen_create_child() is static and should not be exported. tcp4_gso_segment() and tcp6_gso_segment() should be static. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/tcp_fastopen.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net/ipv4/tcp_fastopen.c') diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index ea82fd492c1b..fe77417fc137 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -221,7 +221,6 @@ static bool tcp_fastopen_create_child(struct sock *sk, WARN_ON(req->sk == NULL); return true; } -EXPORT_SYMBOL(tcp_fastopen_create_child); static bool tcp_fastopen_queue_check(struct sock *sk) { -- cgit v1.2.3 From 13854e5a60461daee08ce99842b7f4d37553d911 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sun, 15 Mar 2015 21:12:16 -0700 Subject: inet: add proper refcounting to request sock reqsk_put() is the generic function that should be used to release a refcount (and automatically call reqsk_free()) reqsk_free() might be called if refcount is known to be 0 or undefined. refcnt is set to one in inet_csk_reqsk_queue_add() As request socks are not yet in global ehash table, I added temporary debugging checks in reqsk_put() and reqsk_free() Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/tcp_fastopen.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/ipv4/tcp_fastopen.c') diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index fe77417fc137..84381319e1bc 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -253,7 +253,7 @@ static bool tcp_fastopen_queue_check(struct sock *sk) fastopenq->rskq_rst_head = req1->dl_next; fastopenq->qlen--; spin_unlock(&fastopenq->lock); - reqsk_free(req1); + reqsk_put(req1); } return true; } -- cgit v1.2.3 From 9439ce00f208d95703a6725e4ea986dd90e37ffd Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 17 Mar 2015 18:32:29 -0700 Subject: tcp: rename struct tcp_request_sock listener The listener field in struct tcp_request_sock is a pointer back to the listener. We now have req->rsk_listener, so TCP only needs one boolean and not a full pointer. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/tcp_fastopen.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) (limited to 'net/ipv4/tcp_fastopen.c') diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 84381319e1bc..186fd394ec0a 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -155,12 +155,7 @@ static bool tcp_fastopen_create_child(struct sock *sk, tp = tcp_sk(child); tp->fastopen_rsk = req; - /* Do a hold on the listner sk so that if the listener is being - * closed, the child that has been accepted can live on and still - * access listen_lock. - */ - sock_hold(sk); - tcp_rsk(req)->listener = sk; + tcp_rsk(req)->tfo_listener = true; /* RFC1323: The window in SYN & SYN/ACK segments is never * scaled. So correct it appropriately. -- cgit v1.2.3 From 0470c8ca1d57927f2cc3e1d5add1fb2834609447 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 17 Mar 2015 18:32:31 -0700 Subject: inet: fix request sock refcounting While testing last patch series, I found req sock refcounting was wrong. We must set skc_refcnt to 1 for all request socks added in hashes, but also on request sockets created by FastOpen or syncookies. It is tricky because we need to defer this initialization so that future RCU lookups do not try to take a refcount on a not yet fully initialized request socket. Also get rid of ireq_refcnt alias. Signed-off-by: Eric Dumazet Fixes: 13854e5a6046 ("inet: add proper refcounting to request sock") Signed-off-by: David S. Miller --- net/ipv4/tcp_fastopen.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net/ipv4/tcp_fastopen.c') diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 186fd394ec0a..82e375a0cbcf 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -169,6 +169,7 @@ static bool tcp_fastopen_create_child(struct sock *sk, inet_csk_reset_xmit_timer(child, ICSK_TIME_RETRANS, TCP_TIMEOUT_INIT, TCP_RTO_MAX); + atomic_set(&req->rsk_refcnt, 1); /* Add the child socket directly into the accept queue */ inet_csk_reqsk_queue_add(sk, req, child); -- cgit v1.2.3 From fa76ce7328b289b6edd476e24eb52fd634261720 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 19 Mar 2015 19:04:20 -0700 Subject: inet: get rid of central tcp/dccp listener timer One of the major issue for TCP is the SYNACK rtx handling, done by inet_csk_reqsk_queue_prune(), fired by the keepalive timer of a TCP_LISTEN socket. This function runs for awful long times, with socket lock held, meaning that other cpus needing this lock have to spin for hundred of ms. SYNACK are sent in huge bursts, likely to cause severe drops anyway. This model was OK 15 years ago when memory was very tight. We now can afford to have a timer per request sock. Timer invocations no longer need to lock the listener, and can be run from all cpus in parallel. With following patch increasing somaxconn width to 32 bits, I tested a listener with more than 4 million active request sockets, and a steady SYNFLOOD of ~200,000 SYN per second. Host was sending ~830,000 SYNACK per second. This is ~100 times more what we could achieve before this patch. Later, we will get rid of the listener hash and use ehash instead. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/tcp_fastopen.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/ipv4/tcp_fastopen.c') diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 82e375a0cbcf..2eb887ec0ce3 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -240,7 +240,7 @@ static bool tcp_fastopen_queue_check(struct sock *sk) struct request_sock *req1; spin_lock(&fastopenq->lock); req1 = fastopenq->rskq_rst_head; - if ((req1 == NULL) || time_after(req1->expires, jiffies)) { + if (!req1 || time_after(req1->rsk_timer.expires, jiffies)) { spin_unlock(&fastopenq->lock); NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPFASTOPENLISTENOVERFLOW); -- cgit v1.2.3 From 51456b2914a34d16b1255b7c55d5cbf6a681d306 Mon Sep 17 00:00:00 2001 From: Ian Morris Date: Fri, 3 Apr 2015 09:17:26 +0100 Subject: ipv4: coding style: comparison for equality with NULL The ipv4 code uses a mixture of coding styles. In some instances check for NULL pointer is done as x == NULL and sometimes as !x. !x is preferred according to checkpatch and this patch makes the code consistent by adopting the latter form. No changes detected by objdiff. Signed-off-by: Ian Morris Signed-off-by: David S. Miller --- net/ipv4/tcp_fastopen.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'net/ipv4/tcp_fastopen.c') diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 2eb887ec0ce3..5da55e2b5cd2 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -141,7 +141,7 @@ static bool tcp_fastopen_create_child(struct sock *sk, req->sk = NULL; child = inet_csk(sk)->icsk_af_ops->syn_recv_sock(sk, skb, req, NULL); - if (child == NULL) + if (!child) return false; spin_lock(&queue->fastopenq->lock); @@ -214,7 +214,7 @@ static bool tcp_fastopen_create_child(struct sock *sk, sk->sk_data_ready(sk); bh_unlock_sock(child); sock_put(child); - WARN_ON(req->sk == NULL); + WARN_ON(!req->sk); return true; } @@ -233,7 +233,7 @@ static bool tcp_fastopen_queue_check(struct sock *sk) * temporarily vs a server not supporting Fast Open at all. */ fastopenq = inet_csk(sk)->icsk_accept_queue.fastopenq; - if (fastopenq == NULL || fastopenq->max_qlen == 0) + if (!fastopenq || fastopenq->max_qlen == 0) return false; if (fastopenq->qlen >= fastopenq->max_qlen) { -- cgit v1.2.3 From 7f9b838b71eb78a27de27a12ca5de8542fac3115 Mon Sep 17 00:00:00 2001 From: Daniel Lee Date: Mon, 6 Apr 2015 14:37:26 -0700 Subject: tcp: RFC7413 option support for Fast Open server Fast Open has been using the experimental option with a magic number (RFC6994) to request and grant Fast Open cookies. This patch enables the server to support the official IANA option 34 in RFC7413 in addition. The change has passed all existing Fast Open tests with both old and new options at Google. Signed-off-by: Daniel Lee Signed-off-by: Yuchung Cheng Signed-off-by: Neal Cardwell Signed-off-by: David S. Miller --- net/ipv4/tcp_fastopen.c | 1 + 1 file changed, 1 insertion(+) (limited to 'net/ipv4/tcp_fastopen.c') diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 5da55e2b5cd2..e3d87aca6be8 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -303,6 +303,7 @@ fastopen: } else if (foc->len > 0) /* Client presents an invalid cookie */ NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPFASTOPENPASSIVEFAIL); + valid_foc.exp = foc->exp; *foc = valid_foc; return false; } -- cgit v1.2.3