From 317a76f9a44b437d6301718f4e5d08bd93f98da7 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 23 Jun 2005 12:19:55 -0700 Subject: [TCP]: Add pluggable congestion control algorithm infrastructure. Allow TCP to have multiple pluggable congestion control algorithms. Algorithms are defined by a set of operations and can be built in or modules. The legacy "new RENO" algorithm is used as a starting point and fallback. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/ipv4/tcp_ipv4.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net/ipv4/tcp_ipv4.c') diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 2d41d5d6ad19..9122814c13ad 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2048,6 +2048,7 @@ static int tcp_v4_init_sock(struct sock *sk) tp->mss_cache_std = tp->mss_cache = 536; tp->reordering = sysctl_tcp_reordering; + tp->ca_ops = &tcp_reno; sk->sk_state = TCP_CLOSE; @@ -2070,6 +2071,8 @@ int tcp_v4_destroy_sock(struct sock *sk) tcp_clear_xmit_timers(sk); + tcp_cleanup_congestion_control(tp); + /* Cleanup up the write buffer. */ sk_stream_writequeue_purge(sk); -- cgit v1.2.2 From 5f8ef48d240963093451bcf83df89f1a1364f51d Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 23 Jun 2005 20:37:36 -0700 Subject: [TCP]: Allow choosing TCP congestion control via sockopt. Allow using setsockopt to set TCP congestion control to use on a per socket basis. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/ipv4/tcp_ipv4.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/ipv4/tcp_ipv4.c') diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 9122814c13ad..ebf112347a97 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2048,7 +2048,7 @@ static int tcp_v4_init_sock(struct sock *sk) tp->mss_cache_std = tp->mss_cache = 536; tp->reordering = sysctl_tcp_reordering; - tp->ca_ops = &tcp_reno; + tp->ca_ops = &tcp_init_congestion_ops; sk->sk_state = TCP_CLOSE; -- cgit v1.2.2 From c1b4a7e69576d65efc31a8cea0714173c2841244 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 5 Jul 2005 15:24:38 -0700 Subject: [TCP]: Move to new TSO segmenting scheme. Make TSO segment transmit size decisions at send time not earlier. The basic scheme is that we try to build as large a TSO frame as possible when pulling in the user data, but the size of the TSO frame output to the card is determined at transmit time. This is guided by tp->xmit_size_goal. It is always set to a multiple of MSS and tells sendmsg/sendpage how large an SKB to try and build. Later, tcp_write_xmit() and tcp_push_one() chop up the packet if necessary and conditions warrant. These routines can also decide to "defer" in order to wait for more ACKs to arrive and thus allow larger TSO frames to be emitted. A general observation is that TSO elongates the pipe, thus requiring a larger congestion window and larger buffering especially at the sender side. Therefore, it is important that applications 1) get a large enough socket send buffer (this is accomplished by our dynamic send buffer expansion code) 2) do large enough writes. Signed-off-by: David S. Miller --- net/ipv4/tcp_ipv4.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/ipv4/tcp_ipv4.c') diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index ebf112347a97..62f62bb05c2a 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2045,7 +2045,7 @@ static int tcp_v4_init_sock(struct sock *sk) */ tp->snd_ssthresh = 0x7fffffff; /* Infinity */ tp->snd_cwnd_clamp = ~0; - tp->mss_cache_std = tp->mss_cache = 536; + tp->mss_cache = 536; tp->reordering = sysctl_tcp_reordering; tp->ca_ops = &tcp_init_congestion_ops; -- cgit v1.2.2 From ca9334523c853e407da7b3a0bd02f54d0fa59414 Mon Sep 17 00:00:00 2001 From: Heikki Orsila Date: Mon, 8 Aug 2005 14:26:52 -0700 Subject: [IPV4]: Debug cleanup Here's a small patch to cleanup NETDEBUG() use in net/ipv4/ for Linux kernel 2.6.13-rc5. Also weird use of indentation is changed in some places. Signed-off-by: Heikki Orsila Signed-off-by: David S. Miller --- net/ipv4/tcp_ipv4.c | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) (limited to 'net/ipv4/tcp_ipv4.c') diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 62f62bb05c2a..5d91213d34c0 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1494,12 +1494,11 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) * to destinations, already remembered * to the moment of synflood. */ - NETDEBUG(if (net_ratelimit()) \ - printk(KERN_DEBUG "TCP: drop open " - "request from %u.%u." - "%u.%u/%u\n", \ - NIPQUAD(saddr), - ntohs(skb->h.th->source))); + LIMIT_NETDEBUG(printk(KERN_DEBUG "TCP: drop open " + "request from %u.%u." + "%u.%u/%u\n", + NIPQUAD(saddr), + ntohs(skb->h.th->source))); dst_release(dst); goto drop_and_free; } @@ -1627,8 +1626,7 @@ static int tcp_v4_checksum_init(struct sk_buff *skb) skb->nh.iph->daddr, skb->csum)) return 0; - NETDEBUG(if (net_ratelimit()) - printk(KERN_DEBUG "hw tcp v4 csum failed\n")); + LIMIT_NETDEBUG(printk(KERN_DEBUG "hw tcp v4 csum failed\n")); skb->ip_summed = CHECKSUM_NONE; } if (skb->len <= 76) { -- cgit v1.2.2 From d5d283751ef3c05b6766501a46800cbee84959d6 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 23 Aug 2005 10:49:54 -0700 Subject: [TCP]: Document non-trivial locking path in tcp_v{4,6}_get_port(). This trips up a lot of folks reading this code. Put an unlikely() around the port-exhaustion test for good measure. Signed-off-by: David S. Miller --- net/ipv4/tcp_ipv4.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'net/ipv4/tcp_ipv4.c') diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 5d91213d34c0..67c670886c1f 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -242,9 +242,14 @@ static int tcp_v4_get_port(struct sock *sk, unsigned short snum) tcp_port_rover = rover; spin_unlock(&tcp_portalloc_lock); - /* Exhausted local port range during search? */ + /* Exhausted local port range during search? It is not + * possible for us to be holding one of the bind hash + * locks if this test triggers, because if 'remaining' + * drops to zero, we broke out of the do/while loop at + * the top level, not from the 'break;' statement. + */ ret = 1; - if (remaining <= 0) + if (unlikely(remaining <= 0)) goto fail; /* OK, here is the one we will use. HEAD is -- cgit v1.2.2