From 3b885787ea4112eaa80945999ea0901bf742707f Mon Sep 17 00:00:00 2001 From: Neil Horman Date: Mon, 12 Oct 2009 13:26:31 -0700 Subject: net: Generalize socket rx gap / receive queue overflow cmsg Create a new socket level option to report number of queue overflows Recently I augmented the AF_PACKET protocol to report the number of frames lost on the socket receive queue between any two enqueued frames. This value was exported via a SOL_PACKET level cmsg. AFter I completed that work it was requested that this feature be generalized so that any datagram oriented socket could make use of this option. As such I've created this patch, It creates a new SOL_SOCKET level option called SO_RXQ_OVFL, which when enabled exports a SOL_SOCKET level cmsg that reports the nubmer of times the sk_receive_queue overflowed between any two given frames. It also augments the AF_PACKET protocol to take advantage of this new feature (as it previously did not touch sk->sk_drops, which this patch uses to record the overflow count). Tested successfully by me. Notes: 1) Unlike my previous patch, this patch simply records the sk_drops value, which is not a number of drops between packets, but rather a total number of drops. Deltas must be computed in user space. 2) While this patch currently works with datagram oriented protocols, it will also be accepted by non-datagram oriented protocols. I'm not sure if thats agreeable to everyone, but my argument in favor of doing so is that, for those protocols which aren't applicable to this option, sk_drops will always be zero, and reporting no drops on a receive queue that isn't used for those non-participating protocols seems reasonable to me. This also saves us having to code in a per-protocol opt in mechanism. 3) This applies cleanly to net-next assuming that commit 977750076d98c7ff6cbda51858bb5a5894a9d9ab (my af packet cmsg patch) is reverted Signed-off-by: Neil Horman Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv6/raw.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'net/ipv6/raw.c') diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 4f24570b0869..d8375bc7f2d5 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -497,7 +497,7 @@ static int rawv6_recvmsg(struct kiocb *iocb, struct sock *sk, sin6->sin6_scope_id = IP6CB(skb)->iif; } - sock_recv_timestamp(msg, sk, skb); + sock_recv_ts_and_drops(msg, sk, skb); if (np->rxopt.all) datagram_recv_ctl(sk, msg, skb); -- cgit v1.2.3 From 766e9037cc139ee25ed93ee5ad11e1450c4b99f6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 14 Oct 2009 20:40:11 -0700 Subject: net: sk_drops consolidation sock_queue_rcv_skb() can update sk_drops itself, removing need for callers to take care of it. This is more consistent since sock_queue_rcv_skb() also reads sk_drops when queueing a skb. This adds sk_drops managment to many protocols that not cared yet. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv6/raw.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'net/ipv6/raw.c') diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index d8375bc7f2d5..fd737efed96c 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -381,8 +381,7 @@ static inline int rawv6_rcv_skb(struct sock * sk, struct sk_buff * skb) } /* Charge it to the socket. */ - if (sock_queue_rcv_skb(sk,skb)<0) { - atomic_inc(&sk->sk_drops); + if (sock_queue_rcv_skb(sk, skb) < 0) { kfree_skb(skb); return NET_RX_DROP; } -- cgit v1.2.3 From c720c7e8383aff1cb219bddf474ed89d850336e3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 15 Oct 2009 06:30:45 +0000 Subject: inet: rename some inet_sock fields In order to have better cache layouts of struct sock (separate zones for rx/tx paths), we need this preliminary patch. Goal is to transfert fields used at lookup time in the first read-mostly cache line (inside struct sock_common) and move sk_refcnt to a separate cache line (only written by rx path) This patch adds inet_ prefix to daddr, rcv_saddr, dport, num, saddr, sport and id fields. This allows a future patch to define these fields as macros, like sk_refcnt, without name clashes. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv6/raw.c | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) (limited to 'net/ipv6/raw.c') diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index fd737efed96c..52ed7d7f9dab 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -72,7 +72,7 @@ static struct sock *__raw_v6_lookup(struct net *net, struct sock *sk, int is_multicast = ipv6_addr_is_multicast(loc_addr); sk_for_each_from(sk, node) - if (inet_sk(sk)->num == num) { + if (inet_sk(sk)->inet_num == num) { struct ipv6_pinfo *np = inet6_sk(sk); if (!net_eq(sock_net(sk), net)) @@ -298,7 +298,7 @@ static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) dev_put(dev); } - inet->rcv_saddr = inet->saddr = v4addr; + inet->inet_rcv_saddr = inet->inet_saddr = v4addr; ipv6_addr_copy(&np->rcv_saddr, &addr->sin6_addr); if (!(addr_type & IPV6_ADDR_MULTICAST)) ipv6_addr_copy(&np->saddr, &addr->sin6_addr); @@ -415,14 +415,14 @@ int rawv6_rcv(struct sock *sk, struct sk_buff *skb) skb_network_header_len(skb)); if (!csum_ipv6_magic(&ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr, - skb->len, inet->num, skb->csum)) + skb->len, inet->inet_num, skb->csum)) skb->ip_summed = CHECKSUM_UNNECESSARY; } if (!skb_csum_unnecessary(skb)) skb->csum = ~csum_unfold(csum_ipv6_magic(&ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr, skb->len, - inet->num, 0)); + inet->inet_num, 0)); if (inet->hdrincl) { if (skb_checksum_complete(skb)) { @@ -765,8 +765,8 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk, proto = ntohs(sin6->sin6_port); if (!proto) - proto = inet->num; - else if (proto != inet->num) + proto = inet->inet_num; + else if (proto != inet->inet_num) return(-EINVAL); if (proto > 255) @@ -799,7 +799,7 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk, if (sk->sk_state != TCP_ESTABLISHED) return -EDESTADDRREQ; - proto = inet->num; + proto = inet->inet_num; daddr = &np->daddr; fl.fl6_flowlabel = np->flow_label; } @@ -966,7 +966,7 @@ static int do_rawv6_setsockopt(struct sock *sk, int level, int optname, switch (optname) { case IPV6_CHECKSUM: - if (inet_sk(sk)->num == IPPROTO_ICMPV6 && + if (inet_sk(sk)->inet_num == IPPROTO_ICMPV6 && level == IPPROTO_IPV6) { /* * RFC3542 tells that IPV6_CHECKSUM socket @@ -1006,7 +1006,7 @@ static int rawv6_setsockopt(struct sock *sk, int level, int optname, break; case SOL_ICMPV6: - if (inet_sk(sk)->num != IPPROTO_ICMPV6) + if (inet_sk(sk)->inet_num != IPPROTO_ICMPV6) return -EOPNOTSUPP; return rawv6_seticmpfilter(sk, level, optname, optval, optlen); @@ -1029,7 +1029,7 @@ static int compat_rawv6_setsockopt(struct sock *sk, int level, int optname, case SOL_RAW: break; case SOL_ICMPV6: - if (inet_sk(sk)->num != IPPROTO_ICMPV6) + if (inet_sk(sk)->inet_num != IPPROTO_ICMPV6) return -EOPNOTSUPP; return rawv6_seticmpfilter(sk, level, optname, optval, optlen); case SOL_IPV6: @@ -1086,7 +1086,7 @@ static int rawv6_getsockopt(struct sock *sk, int level, int optname, break; case SOL_ICMPV6: - if (inet_sk(sk)->num != IPPROTO_ICMPV6) + if (inet_sk(sk)->inet_num != IPPROTO_ICMPV6) return -EOPNOTSUPP; return rawv6_geticmpfilter(sk, level, optname, optval, optlen); @@ -1109,7 +1109,7 @@ static int compat_rawv6_getsockopt(struct sock *sk, int level, int optname, case SOL_RAW: break; case SOL_ICMPV6: - if (inet_sk(sk)->num != IPPROTO_ICMPV6) + if (inet_sk(sk)->inet_num != IPPROTO_ICMPV6) return -EOPNOTSUPP; return rawv6_geticmpfilter(sk, level, optname, optval, optlen); case SOL_IPV6: @@ -1156,7 +1156,7 @@ static int rawv6_ioctl(struct sock *sk, int cmd, unsigned long arg) static void rawv6_close(struct sock *sk, long timeout) { - if (inet_sk(sk)->num == IPPROTO_RAW) + if (inet_sk(sk)->inet_num == IPPROTO_RAW) ip6_ra_control(sk, -1); ip6mr_sk_done(sk); sk_common_release(sk); @@ -1175,7 +1175,7 @@ static int rawv6_init_sk(struct sock *sk) { struct raw6_sock *rp = raw6_sk(sk); - switch (inet_sk(sk)->num) { + switch (inet_sk(sk)->inet_num) { case IPPROTO_ICMPV6: rp->checksum = 1; rp->offset = 2; @@ -1225,7 +1225,7 @@ static void raw6_sock_seq_show(struct seq_file *seq, struct sock *sp, int i) dest = &np->daddr; src = &np->rcv_saddr; destp = 0; - srcp = inet_sk(sp)->num; + srcp = inet_sk(sp)->inet_num; seq_printf(seq, "%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X " "%02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p %d\n", -- cgit v1.2.3 From 8edf19c2fe028563fc6ea9cb1995b8ee4172d4b6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 15 Oct 2009 00:12:40 +0000 Subject: net: sk_drops consolidation part 2 - skb_kill_datagram() can increment sk->sk_drops itself, not callers. - UDP on IPV4 & IPV6 dropped frames (because of bad checksum or policy checks) increment sk_drops Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv6/raw.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net/ipv6/raw.c') diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 52ed7d7f9dab..cb834ab7f071 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -517,7 +517,6 @@ csum_copy_err: as some normal condition. */ err = (flags&MSG_DONTWAIT) ? -EAGAIN : -EHOSTUNREACH; - atomic_inc(&sk->sk_drops); goto out; } -- cgit v1.2.3 From 13f18aa05f5abe135f47b6417537ae2b2fedc18c Mon Sep 17 00:00:00 2001 From: Eric Paris Date: Thu, 5 Nov 2009 20:44:37 -0800 Subject: net: drop capability from protocol definitions struct can_proto had a capability field which wasn't ever used. It is dropped entirely. struct inet_protosw had a capability field which can be more clearly expressed in the code by just checking if sock->type = SOCK_RAW. Signed-off-by: Eric Paris Acked-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- net/ipv6/raw.c | 1 - 1 file changed, 1 deletion(-) (limited to 'net/ipv6/raw.c') diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index cb834ab7f071..818ef21ba76d 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -1336,7 +1336,6 @@ static struct inet_protosw rawv6_protosw = { .protocol = IPPROTO_IP, /* wild card */ .prot = &rawv6_prot, .ops = &inet6_sockraw_ops, - .capability = CAP_NET_RAW, .no_check = UDP_CSUM_DEFAULT, .flags = INET_PROTOSW_REUSE, }; -- cgit v1.2.3 From fd5c00276127661f12e0315e3bbec41a1c0be376 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 6 Nov 2009 07:01:17 +0000 Subject: ipv6: avoid dev_hold()/dev_put() in rawv6_bind() Using RCU helps not touching device refcount in rawv6_bind() Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv6/raw.c | 23 +++++++++++------------ 1 file changed, 11 insertions(+), 12 deletions(-) (limited to 'net/ipv6/raw.c') diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 818ef21ba76d..926ce8eeffaf 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -249,7 +249,7 @@ static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) /* Raw sockets are IPv6 only */ if (addr_type == IPV6_ADDR_MAPPED) - return(-EADDRNOTAVAIL); + return -EADDRNOTAVAIL; lock_sock(sk); @@ -257,6 +257,7 @@ static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) if (sk->sk_state != TCP_CLOSE) goto out; + rcu_read_lock(); /* Check if the address belongs to the host. */ if (addr_type != IPV6_ADDR_ANY) { struct net_device *dev = NULL; @@ -272,13 +273,13 @@ static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) /* Binding to link-local address requires an interface */ if (!sk->sk_bound_dev_if) - goto out; + goto out_unlock; - dev = dev_get_by_index(sock_net(sk), sk->sk_bound_dev_if); - if (!dev) { - err = -ENODEV; - goto out; - } + err = -ENODEV; + dev = dev_get_by_index_rcu(sock_net(sk), + sk->sk_bound_dev_if); + if (!dev) + goto out_unlock; } /* ipv4 addr of the socket is invalid. Only the @@ -289,13 +290,9 @@ static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) err = -EADDRNOTAVAIL; if (!ipv6_chk_addr(sock_net(sk), &addr->sin6_addr, dev, 0)) { - if (dev) - dev_put(dev); - goto out; + goto out_unlock; } } - if (dev) - dev_put(dev); } inet->inet_rcv_saddr = inet->inet_saddr = v4addr; @@ -303,6 +300,8 @@ static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) if (!(addr_type & IPV6_ADDR_MULTICAST)) ipv6_addr_copy(&np->saddr, &addr->sin6_addr); err = 0; +out_unlock: + rcu_read_unlock(); out: release_sock(sk); return err; -- cgit v1.2.3