summaryrefslogtreecommitdiffstats
path: root/net/ipv4/udp.c
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2014-08-06 09:38:14 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2014-08-06 09:38:14 -0700
commitae045e2455429c418a418a3376301a9e5753a0a8 (patch)
treeb445bdeecd3f38aa0d0a29c9585cee49e4ccb0f1 /net/ipv4/udp.c
parentf4f142ed4ef835709c7e6d12eaca10d190bcebed (diff)
parentd247b6ab3ce6dd43665780865ec5fa145d9ab6bd (diff)
downloadlinux-ae045e2455429c418a418a3376301a9e5753a0a8.tar.gz
linux-ae045e2455429c418a418a3376301a9e5753a0a8.tar.bz2
linux-ae045e2455429c418a418a3376301a9e5753a0a8.zip
Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next
Pull networking updates from David Miller: "Highlights: 1) Steady transitioning of the BPF instructure to a generic spot so all kernel subsystems can make use of it, from Alexei Starovoitov. 2) SFC driver supports busy polling, from Alexandre Rames. 3) Take advantage of hash table in UDP multicast delivery, from David Held. 4) Lighten locking, in particular by getting rid of the LRU lists, in inet frag handling. From Florian Westphal. 5) Add support for various RFC6458 control messages in SCTP, from Geir Ola Vaagland. 6) Allow to filter bridge forwarding database dumps by device, from Jamal Hadi Salim. 7) virtio-net also now supports busy polling, from Jason Wang. 8) Some low level optimization tweaks in pktgen from Jesper Dangaard Brouer. 9) Add support for ipv6 address generation modes, so that userland can have some input into the process. From Jiri Pirko. 10) Consolidate common TCP connection request code in ipv4 and ipv6, from Octavian Purdila. 11) New ARP packet logger in netfilter, from Pablo Neira Ayuso. 12) Generic resizable RCU hash table, with intial users in netlink and nftables. From Thomas Graf. 13) Maintain a name assignment type so that userspace can see where a network device name came from (enumerated by kernel, assigned explicitly by userspace, etc.) From Tom Gundersen. 14) Automatic flow label generation on transmit in ipv6, from Tom Herbert. 15) New packet timestamping facilities from Willem de Bruijn, meant to assist in measuring latencies going into/out-of the packet scheduler, latency from TCP data transmission to ACK, etc" * git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1536 commits) cxgb4 : Disable recursive mailbox commands when enabling vi net: reduce USB network driver config options. tg3: Modify tg3_tso_bug() to handle multiple TX rings amd-xgbe: Perform phy connect/disconnect at dev open/stop amd-xgbe: Use dma_set_mask_and_coherent to set DMA mask net: sun4i-emac: fix memory leak on bad packet sctp: fix possible seqlock seadlock in sctp_packet_transmit() Revert "net: phy: Set the driver when registering an MDIO bus device" cxgb4vf: Turn off SGE RX/TX Callback Timers and interrupts in PCI shutdown routine team: Simplify return path of team_newlink bridge: Update outdated comment on promiscuous mode net-timestamp: ACK timestamp for bytestreams net-timestamp: TCP timestamping net-timestamp: SCHED timestamp on entering packet scheduler net-timestamp: add key to disambiguate concurrent datagrams net-timestamp: move timestamp flags out of sk_flags net-timestamp: extend SCM_TIMESTAMPING ancillary data struct cxgb4i : Move stray CPL definitions to cxgb4 driver tcp: reduce spurious retransmits due to transient SACK reneging qlcnic: Initialize dcbnl_ops before register_netdev ...
Diffstat (limited to 'net/ipv4/udp.c')
-rw-r--r--net/ipv4/udp.c156
1 files changed, 35 insertions, 121 deletions
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 7d5a8661df76..f57c0e4c2326 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -594,27 +594,6 @@ static inline bool __udp_is_mcast_sock(struct net *net, struct sock *sk,
return true;
}
-static inline struct sock *udp_v4_mcast_next(struct net *net, struct sock *sk,
- __be16 loc_port, __be32 loc_addr,
- __be16 rmt_port, __be32 rmt_addr,
- int dif)
-{
- struct hlist_nulls_node *node;
- struct sock *s = sk;
- unsigned short hnum = ntohs(loc_port);
-
- sk_nulls_for_each_from(s, node) {
- if (__udp_is_mcast_sock(net, s,
- loc_port, loc_addr,
- rmt_port, rmt_addr,
- dif, hnum))
- goto found;
- }
- s = NULL;
-found:
- return s;
-}
-
/*
* This routine is called by the ICMP module when it gets some
* sort of error condition. If err < 0 then the socket should
@@ -1588,7 +1567,7 @@ int udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
goto csum_error;
- if (sk_rcvqueues_full(sk, skb, sk->sk_rcvbuf)) {
+ if (sk_rcvqueues_full(sk, sk->sk_rcvbuf)) {
UDP_INC_STATS_BH(sock_net(sk), UDP_MIB_RCVBUFERRORS,
is_udplite);
goto drop;
@@ -1640,6 +1619,8 @@ static void flush_stack(struct sock **stack, unsigned int count,
if (skb1 && udp_queue_rcv_skb(sk, skb1) <= 0)
skb1 = NULL;
+
+ sock_put(sk);
}
if (unlikely(skb1))
kfree_skb(skb1);
@@ -1668,41 +1649,50 @@ static int __udp4_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
struct udp_table *udptable)
{
struct sock *sk, *stack[256 / sizeof(struct sock *)];
- struct udp_hslot *hslot = udp_hashslot(udptable, net, ntohs(uh->dest));
- int dif;
- unsigned int i, count = 0;
+ struct hlist_nulls_node *node;
+ unsigned short hnum = ntohs(uh->dest);
+ struct udp_hslot *hslot = udp_hashslot(udptable, net, hnum);
+ int dif = skb->dev->ifindex;
+ unsigned int count = 0, offset = offsetof(typeof(*sk), sk_nulls_node);
+ unsigned int hash2 = 0, hash2_any = 0, use_hash2 = (hslot->count > 10);
+
+ if (use_hash2) {
+ hash2_any = udp4_portaddr_hash(net, htonl(INADDR_ANY), hnum) &
+ udp_table.mask;
+ hash2 = udp4_portaddr_hash(net, daddr, hnum) & udp_table.mask;
+start_lookup:
+ hslot = &udp_table.hash2[hash2];
+ offset = offsetof(typeof(*sk), __sk_common.skc_portaddr_node);
+ }
spin_lock(&hslot->lock);
- sk = sk_nulls_head(&hslot->head);
- dif = skb->dev->ifindex;
- sk = udp_v4_mcast_next(net, sk, uh->dest, daddr, uh->source, saddr, dif);
- while (sk) {
- stack[count++] = sk;
- sk = udp_v4_mcast_next(net, sk_nulls_next(sk), uh->dest,
- daddr, uh->source, saddr, dif);
- if (unlikely(count == ARRAY_SIZE(stack))) {
- if (!sk)
- break;
- flush_stack(stack, count, skb, ~0);
- count = 0;
+ sk_nulls_for_each_entry_offset(sk, node, &hslot->head, offset) {
+ if (__udp_is_mcast_sock(net, sk,
+ uh->dest, daddr,
+ uh->source, saddr,
+ dif, hnum)) {
+ if (unlikely(count == ARRAY_SIZE(stack))) {
+ flush_stack(stack, count, skb, ~0);
+ count = 0;
+ }
+ stack[count++] = sk;
+ sock_hold(sk);
}
}
- /*
- * before releasing chain lock, we must take a reference on sockets
- */
- for (i = 0; i < count; i++)
- sock_hold(stack[i]);
spin_unlock(&hslot->lock);
+ /* Also lookup *:port if we are using hash2 and haven't done so yet. */
+ if (use_hash2 && hash2 != hash2_any) {
+ hash2 = hash2_any;
+ goto start_lookup;
+ }
+
/*
* do the slow work with no lock held
*/
if (count) {
flush_stack(stack, count, skb, count - 1);
-
- for (i = 0; i < count; i++)
- sock_put(stack[i]);
} else {
kfree_skb(skb);
}
@@ -2526,79 +2516,3 @@ void __init udp_init(void)
sysctl_udp_rmem_min = SK_MEM_QUANTUM;
sysctl_udp_wmem_min = SK_MEM_QUANTUM;
}
-
-struct sk_buff *skb_udp_tunnel_segment(struct sk_buff *skb,
- netdev_features_t features)
-{
- struct sk_buff *segs = ERR_PTR(-EINVAL);
- u16 mac_offset = skb->mac_header;
- int mac_len = skb->mac_len;
- int tnl_hlen = skb_inner_mac_header(skb) - skb_transport_header(skb);
- __be16 protocol = skb->protocol;
- netdev_features_t enc_features;
- int udp_offset, outer_hlen;
- unsigned int oldlen;
- bool need_csum;
-
- oldlen = (u16)~skb->len;
-
- if (unlikely(!pskb_may_pull(skb, tnl_hlen)))
- goto out;
-
- skb->encapsulation = 0;
- __skb_pull(skb, tnl_hlen);
- skb_reset_mac_header(skb);
- skb_set_network_header(skb, skb_inner_network_offset(skb));
- skb->mac_len = skb_inner_network_offset(skb);
- skb->protocol = htons(ETH_P_TEB);
-
- need_csum = !!(skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL_CSUM);
- if (need_csum)
- skb->encap_hdr_csum = 1;
-
- /* segment inner packet. */
- enc_features = skb->dev->hw_enc_features & netif_skb_features(skb);
- segs = skb_mac_gso_segment(skb, enc_features);
- if (!segs || IS_ERR(segs)) {
- skb_gso_error_unwind(skb, protocol, tnl_hlen, mac_offset,
- mac_len);
- goto out;
- }
-
- outer_hlen = skb_tnl_header_len(skb);
- udp_offset = outer_hlen - tnl_hlen;
- skb = segs;
- do {
- struct udphdr *uh;
- int len;
-
- skb_reset_inner_headers(skb);
- skb->encapsulation = 1;
-
- skb->mac_len = mac_len;
-
- skb_push(skb, outer_hlen);
- skb_reset_mac_header(skb);
- skb_set_network_header(skb, mac_len);
- skb_set_transport_header(skb, udp_offset);
- len = skb->len - udp_offset;
- uh = udp_hdr(skb);
- uh->len = htons(len);
-
- if (need_csum) {
- __be32 delta = htonl(oldlen + len);
-
- uh->check = ~csum_fold((__force __wsum)
- ((__force u32)uh->check +
- (__force u32)delta));
- uh->check = gso_make_checksum(skb, ~uh->check);
-
- if (uh->check == 0)
- uh->check = CSUM_MANGLED_0;
- }
-
- skb->protocol = protocol;
- } while ((skb = skb->next));
-out:
- return segs;
-}