From 8465a5fcd1ceba8f2b55121d47b73f4025401490 Mon Sep 17 00:00:00 2001 From: Neil Horman Date: Thu, 17 Apr 2014 15:26:51 -0400 Subject: sctp: add support for busy polling to sctp protocol The busy polling socket option adds support for sockets to busy wait on data arriving on the napi queue from which they have most recently received a frame. Currently only tcp and udp support this feature, but theres no reason sctp can't do so as well. Add it in so appliations can take advantage of it Signed-off-by: Neil Horman CC: Vlad Yasevich CC: "David S. Miller" CC: Daniel Borkmann CC: netdev@vger.kernel.org Acked-by: Vlad Yasevich Acked-by: Daniel Borkmann Signed-off-by: David S. Miller --- net/sctp/socket.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'net/sctp/socket.c') diff --git a/net/sctp/socket.c b/net/sctp/socket.c index fee06b99a4d..e37b2cbbf17 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -71,6 +71,7 @@ #include #include #include +#include #include /* for sa_family_t */ #include @@ -6557,6 +6558,10 @@ static struct sk_buff *sctp_skb_recv_datagram(struct sock *sk, int flags, if (sk->sk_shutdown & RCV_SHUTDOWN) break; + if (sk_can_busy_loop(sk) && + sk_busy_loop(sk, noblock)) + continue; + /* User doesn't want to wait. */ error = -EAGAIN; if (!timeo) -- cgit v1.2.3-70-g09d2 From 122ff243f5f104194750ecbc76d5946dd1eec934 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Mon, 12 May 2014 16:04:53 -0700 Subject: ipv4: make ip_local_reserved_ports per netns ip_local_port_range is already per netns, so should ip_local_reserved_ports be. And since it is none by default we don't actually need it when we don't enable CONFIG_SYSCTL. By the way, rename inet_is_reserved_local_port() to inet_is_local_reserved_port() Cc: "David S. Miller" Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- include/net/ip.h | 14 +++++++++++--- include/net/netns/ipv4.h | 4 ++++ kernel/sysctl.c | 4 ++-- net/ipv4/af_inet.c | 8 +------- net/ipv4/inet_connection_sock.c | 5 +---- net/ipv4/inet_hashtables.c | 2 +- net/ipv4/sysctl_net_ipv4.c | 31 ++++++++++++++----------------- net/ipv4/udp.c | 2 +- net/sctp/socket.c | 5 +++-- 9 files changed, 38 insertions(+), 37 deletions(-) (limited to 'net/sctp/socket.c') diff --git a/include/net/ip.h b/include/net/ip.h index 14c50a1650e..512bcd5daba 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -208,11 +208,19 @@ static inline u64 snmp_fold_field64(void __percpu *mib, int offt, size_t syncp_o void inet_get_local_port_range(struct net *net, int *low, int *high); -extern unsigned long *sysctl_local_reserved_ports; -static inline int inet_is_reserved_local_port(int port) +#if CONFIG_SYSCTL +static inline int inet_is_local_reserved_port(struct net *net, int port) { - return test_bit(port, sysctl_local_reserved_ports); + if (!net->ipv4.sysctl_local_reserved_ports) + return 0; + return test_bit(port, net->ipv4.sysctl_local_reserved_ports); } +#else +static inline int inet_is_local_reserved_port(struct net *net, int port) +{ + return 0; +} +#endif extern int sysctl_ip_nonlocal_bind; diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 2f0cfad6666..aec5e12f9f1 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -84,6 +84,10 @@ struct netns_ipv4 { atomic_t dev_addr_genid; +#ifdef CONFIG_SYSCTL + unsigned long *sysctl_local_reserved_ports; +#endif + #ifdef CONFIG_IP_MROUTE #ifndef CONFIG_IP_MROUTE_MULTIPLE_TABLES struct mr_table *mrt; diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 74f5b580fe3..e36ae4b1572 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -2501,11 +2501,11 @@ int proc_do_large_bitmap(struct ctl_table *table, int write, bool first = 1; size_t left = *lenp; unsigned long bitmap_len = table->maxlen; - unsigned long *bitmap = (unsigned long *) table->data; + unsigned long *bitmap = *(unsigned long **) table->data; unsigned long *tmp_bitmap = NULL; char tr_a[] = { '-', ',', '\n' }, tr_b[] = { ',', '\n', 0 }, c; - if (!bitmap_len || !left || (*ppos && !write)) { + if (!bitmap || !bitmap_len || !left || (*ppos && !write)) { *lenp = 0; return 0; } diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 211c0cc6c3d..279132bcadd 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1705,13 +1705,9 @@ static int __init inet_init(void) BUILD_BUG_ON(sizeof(struct inet_skb_parm) > FIELD_SIZEOF(struct sk_buff, cb)); - sysctl_local_reserved_ports = kzalloc(65536 / 8, GFP_KERNEL); - if (!sysctl_local_reserved_ports) - goto out; - rc = proto_register(&tcp_prot, 1); if (rc) - goto out_free_reserved_ports; + goto out; rc = proto_register(&udp_prot, 1); if (rc) @@ -1821,8 +1817,6 @@ out_unregister_udp_proto: proto_unregister(&udp_prot); out_unregister_tcp_proto: proto_unregister(&tcp_prot); -out_free_reserved_ports: - kfree(sysctl_local_reserved_ports); goto out; } diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 12e502cbfdc..14d02ea905b 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -29,9 +29,6 @@ const char inet_csk_timer_bug_msg[] = "inet_csk BUG: unknown timer value\n"; EXPORT_SYMBOL(inet_csk_timer_bug_msg); #endif -unsigned long *sysctl_local_reserved_ports; -EXPORT_SYMBOL(sysctl_local_reserved_ports); - void inet_get_local_port_range(struct net *net, int *low, int *high) { unsigned int seq; @@ -113,7 +110,7 @@ again: smallest_size = -1; do { - if (inet_is_reserved_local_port(rover)) + if (inet_is_local_reserved_port(net, rover)) goto next_nolock; head = &hashinfo->bhash[inet_bhashfn(net, rover, hashinfo->bhash_size)]; diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 8b9cf279450..83331f1b86a 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -500,7 +500,7 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, local_bh_disable(); for (i = 1; i <= remaining; i++) { port = low + (i + offset) % remaining; - if (inet_is_reserved_local_port(port)) + if (inet_is_local_reserved_port(net, port)) continue; head = &hinfo->bhash[inet_bhashfn(net, port, hinfo->bhash_size)]; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index a33b9fbc1d8..79a007c5255 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -436,13 +436,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "ip_local_reserved_ports", - .data = NULL, /* initialized in sysctl_ipv4_init */ - .maxlen = 65536, - .mode = 0644, - .proc_handler = proc_do_large_bitmap, - }, { .procname = "igmp_max_memberships", .data = &sysctl_igmp_max_memberships, @@ -824,6 +817,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = ipv4_local_port_range, }, + { + .procname = "ip_local_reserved_ports", + .data = &init_net.ipv4.sysctl_local_reserved_ports, + .maxlen = 65536, + .mode = 0644, + .proc_handler = proc_do_large_bitmap, + }, { .procname = "ip_no_pmtu_disc", .data = &init_net.ipv4.sysctl_ip_no_pmtu_disc, @@ -876,8 +876,14 @@ static __net_init int ipv4_sysctl_init_net(struct net *net) if (net->ipv4.ipv4_hdr == NULL) goto err_reg; + net->ipv4.sysctl_local_reserved_ports = kzalloc(65536 / 8, GFP_KERNEL); + if (!net->ipv4.sysctl_local_reserved_ports) + goto err_ports; + return 0; +err_ports: + unregister_net_sysctl_table(net->ipv4.ipv4_hdr); err_reg: if (!net_eq(net, &init_net)) kfree(table); @@ -889,6 +895,7 @@ static __net_exit void ipv4_sysctl_exit_net(struct net *net) { struct ctl_table *table; + kfree(net->ipv4.sysctl_local_reserved_ports); table = net->ipv4.ipv4_hdr->ctl_table_arg; unregister_net_sysctl_table(net->ipv4.ipv4_hdr); kfree(table); @@ -902,16 +909,6 @@ static __net_initdata struct pernet_operations ipv4_sysctl_ops = { static __init int sysctl_ipv4_init(void) { struct ctl_table_header *hdr; - struct ctl_table *i; - - for (i = ipv4_table; i->procname; i++) { - if (strcmp(i->procname, "ip_local_reserved_ports") == 0) { - i->data = sysctl_local_reserved_ports; - break; - } - } - if (!i->procname) - return -EINVAL; hdr = register_net_sysctl(&init_net, "net/ipv4", ipv4_table); if (hdr == NULL) diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 54ea0a3a48f..6729ea97a59 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -246,7 +246,7 @@ int udp_lib_get_port(struct sock *sk, unsigned short snum, do { if (low <= snum && snum <= high && !test_bit(snum >> udptable->log, bitmap) && - !inet_is_reserved_local_port(snum)) + !inet_is_local_reserved_port(net, snum)) goto found; snum += rand; } while (snum != first); diff --git a/net/sctp/socket.c b/net/sctp/socket.c index e37b2cbbf17..2af76eaba8f 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -5946,8 +5946,9 @@ static long sctp_get_port_local(struct sock *sk, union sctp_addr *addr) /* Search for an available port. */ int low, high, remaining, index; unsigned int rover; + struct net *net = sock_net(sk); - inet_get_local_port_range(sock_net(sk), &low, &high); + inet_get_local_port_range(net, &low, &high); remaining = (high - low) + 1; rover = prandom_u32() % remaining + low; @@ -5955,7 +5956,7 @@ static long sctp_get_port_local(struct sock *sk, union sctp_addr *addr) rover++; if ((rover < low) || (rover > high)) rover = low; - if (inet_is_reserved_local_port(rover)) + if (inet_is_local_reserved_port(net, rover)) continue; index = sctp_phashfn(sock_net(sk), rover); head = &sctp_port_hashtable[index]; -- cgit v1.2.3-70-g09d2 From 28448b80456feafe07e2d05b6363b00f61f6171e Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Fri, 23 May 2014 08:47:19 -0700 Subject: net: Split sk_no_check into sk_no_check_{rx,tx} Define separate fields in the sock structure for configuring disabling checksums in both TX and RX-- sk_no_check_tx and sk_no_check_rx. The SO_NO_CHECK socket option only affects sk_no_check_tx. Also, removed UDP_CSUM_* defines since they are no longer necessary. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- drivers/scsi/iscsi_tcp.c | 2 +- include/net/sock.h | 6 ++++-- include/net/udp.h | 9 --------- net/appletalk/ddp.c | 2 +- net/core/sock.c | 4 ++-- net/decnet/af_decnet.c | 2 +- net/ipv4/udp.c | 2 +- net/ipv6/udp.c | 6 +++--- net/ipx/af_ipx.c | 2 +- net/ipx/ipx_route.c | 3 ++- net/l2tp/l2tp_core.c | 4 ++-- net/l2tp/l2tp_netlink.c | 3 +-- net/sctp/socket.c | 3 ++- 13 files changed, 21 insertions(+), 27 deletions(-) (limited to 'net/sctp/socket.c') diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c index 11854845393..a669f2d11c3 100644 --- a/drivers/scsi/iscsi_tcp.c +++ b/drivers/scsi/iscsi_tcp.c @@ -244,7 +244,7 @@ iscsi_sw_tcp_conn_restore_callbacks(struct iscsi_conn *conn) sk->sk_data_ready = tcp_sw_conn->old_data_ready; sk->sk_state_change = tcp_sw_conn->old_state_change; sk->sk_write_space = tcp_sw_conn->old_write_space; - sk->sk_no_check = 0; + sk->sk_no_check_tx = 0; write_unlock_bh(&sk->sk_callback_lock); } diff --git a/include/net/sock.h b/include/net/sock.h index 21569cf456e..07b7fcd60d8 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -243,7 +243,8 @@ struct cg_proto; * @sk_sndbuf: size of send buffer in bytes * @sk_flags: %SO_LINGER (l_onoff), %SO_BROADCAST, %SO_KEEPALIVE, * %SO_OOBINLINE settings, %SO_TIMESTAMPING settings - * @sk_no_check: %SO_NO_CHECK setting, whether or not checkup packets + * @sk_no_check_tx: %SO_NO_CHECK setting, set checksum in TX packets + * @sk_no_check_rx: allow zero checksum in RX packets * @sk_route_caps: route capabilities (e.g. %NETIF_F_TSO) * @sk_route_nocaps: forbidden route capabilities (e.g NETIF_F_GSO_MASK) * @sk_gso_type: GSO type (e.g. %SKB_GSO_TCPV4) @@ -371,7 +372,8 @@ struct sock { struct sk_buff_head sk_write_queue; kmemcheck_bitfield_begin(flags); unsigned int sk_shutdown : 2, - sk_no_check : 2, + sk_no_check_tx : 1, + sk_no_check_rx : 1, sk_userlocks : 4, sk_protocol : 8, sk_type : 16; diff --git a/include/net/udp.h b/include/net/udp.h index a24f0f3e107..5eb86874bcd 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -95,15 +95,6 @@ static inline struct udp_hslot *udp_hashslot2(struct udp_table *table, return &table->hash2[hash & table->mask]; } -/* Note: this must match 'valbool' in sock_setsockopt */ -#define UDP_CSUM_NOXMIT 1 - -/* Used by SunRPC/xprt layer. */ -#define UDP_CSUM_NORCV 2 - -/* Default, as per the RFC, is to always do csums. */ -#define UDP_CSUM_DEFAULT 0 - extern struct proto udp_prot; extern atomic_long_t udp_memory_allocated; diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c index 786ee2f83d5..01a1082e02b 100644 --- a/net/appletalk/ddp.c +++ b/net/appletalk/ddp.c @@ -1669,7 +1669,7 @@ static int atalk_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr goto out; } - if (sk->sk_no_check == 1) + if (sk->sk_no_check_tx) ddp->deh_sum = 0; else ddp->deh_sum = atalk_checksum(skb, len + sizeof(*ddp)); diff --git a/net/core/sock.c b/net/core/sock.c index 664ee4295b6..026e01f7027 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -784,7 +784,7 @@ set_rcvbuf: break; case SO_NO_CHECK: - sk->sk_no_check = valbool; + sk->sk_no_check_tx = valbool; break; case SO_PRIORITY: @@ -1064,7 +1064,7 @@ int sock_getsockopt(struct socket *sock, int level, int optname, break; case SO_NO_CHECK: - v.val = sk->sk_no_check; + v.val = sk->sk_no_check_tx; break; case SO_PRIORITY: diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c index 4c04848953b..ae011b46c07 100644 --- a/net/decnet/af_decnet.c +++ b/net/decnet/af_decnet.c @@ -481,7 +481,7 @@ static struct sock *dn_alloc_sock(struct net *net, struct socket *sock, gfp_t gf sk->sk_backlog_rcv = dn_nsp_backlog_rcv; sk->sk_destruct = dn_destruct; - sk->sk_no_check = 1; + sk->sk_no_check_tx = 1; sk->sk_family = PF_DECnet; sk->sk_protocol = 0; sk->sk_allocation = gfp; diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 590532a7bd2..12c6175b29c 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -785,7 +785,7 @@ static int udp_send_skb(struct sk_buff *skb, struct flowi4 *fl4) if (is_udplite) /* UDP-Lite */ csum = udplite_csum(skb); - else if (sk->sk_no_check == UDP_CSUM_NOXMIT) { /* UDP csum disabled */ + else if (sk->sk_no_check_tx) { /* UDP csum disabled */ skb->ip_summed = CHECKSUM_NONE; goto send; diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index c7ed47bde54..b8db453133a 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -797,7 +797,7 @@ static int __udp6_lib_mcast_deliver(struct net *net, struct sk_buff *skb, /* If zero checksum and sk_no_check is not on for * the socket then skip it. */ - if (uh->check || sk->sk_no_check) + if (uh->check || sk->sk_no_check_rx) stack[count++] = sk; sk = udp_v6_mcast_next(net, sk_nulls_next(sk), uh->dest, daddr, @@ -887,7 +887,7 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, if (sk != NULL) { int ret; - if (!uh->check && !sk->sk_no_check) { + if (!uh->check && !sk->sk_no_check_rx) { sock_put(sk); udp6_csum_zero_error(skb); goto csum_error; @@ -1037,7 +1037,7 @@ static int udp_v6_push_pending_frames(struct sock *sk) if (is_udplite) csum = udplite_csum_outgoing(sk, skb); - else if (sk->sk_no_check == UDP_CSUM_NOXMIT) { /* UDP csum disabled */ + else if (sk->sk_no_check_tx) { /* UDP csum disabled */ skb->ip_summed = CHECKSUM_NONE; goto send; } else if (skb->ip_summed == CHECKSUM_PARTIAL) { /* UDP hardware csum */ diff --git a/net/ipx/af_ipx.c b/net/ipx/af_ipx.c index 41e4e93cb3a..91729b807c7 100644 --- a/net/ipx/af_ipx.c +++ b/net/ipx/af_ipx.c @@ -1353,7 +1353,7 @@ static int ipx_create(struct net *net, struct socket *sock, int protocol, sk_refcnt_debug_inc(sk); sock_init_data(sock, sk); - sk->sk_no_check = 1; /* Checksum off by default */ + sk->sk_no_check_tx = 1; /* Checksum off by default */ sock->ops = &ipx_dgram_ops; rc = 0; out: diff --git a/net/ipx/ipx_route.c b/net/ipx/ipx_route.c index c1f03185c5e..67e7ad3d46b 100644 --- a/net/ipx/ipx_route.c +++ b/net/ipx/ipx_route.c @@ -236,7 +236,8 @@ int ipxrtr_route_packet(struct sock *sk, struct sockaddr_ipx *usipx, } /* Apply checksum. Not allowed on 802.3 links. */ - if (sk->sk_no_check || intrfc->if_dlink_type == htons(IPX_FRAME_8023)) + if (sk->sk_no_check_tx || + intrfc->if_dlink_type == htons(IPX_FRAME_8023)) ipx->ipx_checksum = htons(0xFFFF); else ipx->ipx_checksum = ipx_cksum(ipx, len + sizeof(struct ipxhdr)); diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index ed0716a075b..a1186105f53 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -1188,7 +1188,7 @@ int l2tp_xmit_skb(struct l2tp_session *session, struct sk_buff *skb, int hdr_len l2tp_xmit_ipv6_csum(sk, skb, udp_len); else #endif - if (sk->sk_no_check == UDP_CSUM_NOXMIT) + if (sk->sk_no_check_tx) skb->ip_summed = CHECKSUM_NONE; else if ((skb_dst(skb) && skb_dst(skb)->dev) && (!(skb_dst(skb)->dev->features & NETIF_F_V4_CSUM))) { @@ -1463,7 +1463,7 @@ static int l2tp_tunnel_sock_create(struct net *net, } if (!cfg->use_udp_checksums) - sock->sk->sk_no_check = UDP_CSUM_NOXMIT; + sock->sk->sk_no_check_tx = 1; break; diff --git a/net/l2tp/l2tp_netlink.c b/net/l2tp/l2tp_netlink.c index bd7387adea9..f3d331bdd70 100644 --- a/net/l2tp/l2tp_netlink.c +++ b/net/l2tp/l2tp_netlink.c @@ -297,8 +297,7 @@ static int l2tp_nl_tunnel_send(struct sk_buff *skb, u32 portid, u32 seq, int fla case L2TP_ENCAPTYPE_UDP: if (nla_put_u16(skb, L2TP_ATTR_UDP_SPORT, ntohs(inet->inet_sport)) || nla_put_u16(skb, L2TP_ATTR_UDP_DPORT, ntohs(inet->inet_dport)) || - nla_put_u8(skb, L2TP_ATTR_UDP_CSUM, - (sk->sk_no_check != UDP_CSUM_NOXMIT))) + nla_put_u8(skb, L2TP_ATTR_UDP_CSUM, !sk->sk_no_check_tx)) goto nla_put_failure; /* NOBREAK */ case L2TP_ENCAPTYPE_IP: diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 2af76eaba8f..42989968940 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -6946,7 +6946,8 @@ void sctp_copy_sock(struct sock *newsk, struct sock *sk, newsk->sk_type = sk->sk_type; newsk->sk_bound_dev_if = sk->sk_bound_dev_if; newsk->sk_flags = sk->sk_flags; - newsk->sk_no_check = sk->sk_no_check; + newsk->sk_no_check_tx = sk->sk_no_check_tx; + newsk->sk_no_check_rx = sk->sk_no_check_rx; newsk->sk_reuse = sk->sk_reuse; newsk->sk_shutdown = sk->sk_shutdown; -- cgit v1.2.3-70-g09d2