X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=datapath%2Factions.c;h=208f260e0059ef1da9b7af8b5c0af63dda54af05;hb=4b9f923273fb79079fcfa15350299b71a0062969;hp=a28e98662db314b23486b48d1f321c1b1a7f1aed;hpb=077257b83c68a36ea86f2d21c8395f60df710c21;p=openvswitch diff --git a/datapath/actions.c b/datapath/actions.c index a28e9866..208f260e 100644 --- a/datapath/actions.c +++ b/datapath/actions.c @@ -1,13 +1,21 @@ /* - * Distributed under the terms of the GNU GPL version 2. - * Copyright (c) 2007, 2008, 2009, 2010, 2011 Nicira Networks. + * Copyright (c) 2007-2012 Nicira, Inc. * - * Significant portions of this file may be copied from parts of the Linux - * kernel, by Linus Torvalds and others. + * This program is free software; you can redistribute it and/or + * modify it under the terms of version 2 of the GNU General Public + * License as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + * 02110-1301, USA */ -/* Functions for executing flow actions. */ - #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include @@ -19,11 +27,10 @@ #include #include #include -#include #include #include +#include -#include "actions.h" #include "checksum.h" #include "datapath.h" #include "vlan.h" @@ -43,8 +50,7 @@ static int make_writable(struct sk_buff *skb, int write_len) /* remove VLAN header from packet and update csum accrodingly. */ static int __pop_vlan_tci(struct sk_buff *skb, __be16 *current_tci) { - struct ethhdr *eh; - struct vlan_ethhdr *veth; + struct vlan_hdr *vhdr; int err; err = make_writable(skb, VLAN_ETH_HLEN); @@ -55,15 +61,15 @@ static int __pop_vlan_tci(struct sk_buff *skb, __be16 *current_tci) skb->csum = csum_sub(skb->csum, csum_partial(skb->data + ETH_HLEN, VLAN_HLEN, 0)); - veth = (struct vlan_ethhdr *) skb->data; - *current_tci = veth->h_vlan_TCI; + vhdr = (struct vlan_hdr *)(skb->data + ETH_HLEN); + *current_tci = vhdr->h_vlan_TCI; memmove(skb->data + VLAN_HLEN, skb->data, 2 * ETH_ALEN); + __skb_pull(skb, VLAN_HLEN); - eh = (struct ethhdr *)__skb_pull(skb, VLAN_HLEN); - - skb->protocol = eh->h_proto; + vlan_set_encap_proto(skb, vhdr); skb->mac_header += VLAN_HLEN; + skb_reset_mac_len(skb); return 0; } @@ -77,7 +83,7 @@ static int pop_vlan(struct sk_buff *skb) vlan_set_tci(skb, 0); } else { if (unlikely(skb->protocol != htons(ETH_P_8021Q) || - skb->len < VLAN_ETH_HLEN)) + skb->len < VLAN_ETH_HLEN)) return 0; err = __pop_vlan_tci(skb, &tci); @@ -86,7 +92,7 @@ static int pop_vlan(struct sk_buff *skb) } /* move next vlan tag to hw accel tag */ if (likely(skb->protocol != htons(ETH_P_8021Q) || - skb->len < VLAN_ETH_HLEN)) + skb->len < VLAN_ETH_HLEN)) return 0; err = __pop_vlan_tci(skb, &tci); @@ -97,7 +103,7 @@ static int pop_vlan(struct sk_buff *skb) return 0; } -static int push_vlan(struct sk_buff *skb, __be16 new_tci) +static int push_vlan(struct sk_buff *skb, const struct ovs_action_push_vlan *vlan) { if (unlikely(vlan_tx_tag_present(skb))) { u16 current_tag; @@ -113,117 +119,145 @@ static int push_vlan(struct sk_buff *skb, __be16 new_tci) + ETH_HLEN, VLAN_HLEN, 0)); } - __vlan_hwaccel_put_tag(skb, ntohs(new_tci)); + __vlan_hwaccel_put_tag(skb, ntohs(vlan->vlan_tci) & ~VLAN_TAG_PRESENT); return 0; } -static bool is_ip(struct sk_buff *skb) +static int set_eth_addr(struct sk_buff *skb, + const struct ovs_key_ethernet *eth_key) { - return (OVS_CB(skb)->flow->key.eth.type == htons(ETH_P_IP) && - skb->transport_header > skb->network_header); + int err; + err = make_writable(skb, ETH_HLEN); + if (unlikely(err)) + return err; + + memcpy(eth_hdr(skb)->h_source, eth_key->eth_src, ETH_ALEN); + memcpy(eth_hdr(skb)->h_dest, eth_key->eth_dst, ETH_ALEN); + + return 0; } -static __sum16 *get_l4_checksum(struct sk_buff *skb) +static void set_ip_addr(struct sk_buff *skb, struct iphdr *nh, + __be32 *addr, __be32 new_addr) { - u8 nw_proto = OVS_CB(skb)->flow->key.ip.proto; int transport_len = skb->len - skb_transport_offset(skb); - if (nw_proto == IPPROTO_TCP) { + + if (nh->protocol == IPPROTO_TCP) { if (likely(transport_len >= sizeof(struct tcphdr))) - return &tcp_hdr(skb)->check; - } else if (nw_proto == IPPROTO_UDP) { - if (likely(transport_len >= sizeof(struct udphdr))) - return &udp_hdr(skb)->check; + inet_proto_csum_replace4(&tcp_hdr(skb)->check, skb, + *addr, new_addr, 1); + } else if (nh->protocol == IPPROTO_UDP) { + if (likely(transport_len >= sizeof(struct udphdr))) { + struct udphdr *uh = udp_hdr(skb); + + if (uh->check || + get_ip_summed(skb) == OVS_CSUM_PARTIAL) { + inet_proto_csum_replace4(&uh->check, skb, + *addr, new_addr, 1); + if (!uh->check) + uh->check = CSUM_MANGLED_0; + } + } } - return NULL; + + csum_replace4(&nh->check, *addr, new_addr); + skb_clear_rxhash(skb); + *addr = new_addr; } -static int set_nw_addr(struct sk_buff *skb, const struct nlattr *a) +static void set_ip_ttl(struct sk_buff *skb, struct iphdr *nh, u8 new_ttl) +{ + csum_replace2(&nh->check, htons(nh->ttl << 8), htons(new_ttl << 8)); + nh->ttl = new_ttl; +} + +static int set_ipv4(struct sk_buff *skb, const struct ovs_key_ipv4 *ipv4_key) { - __be32 new_nwaddr = nla_get_be32(a); struct iphdr *nh; - __sum16 *check; - __be32 *nwaddr; int err; - if (unlikely(!is_ip(skb))) - return 0; - err = make_writable(skb, skb_network_offset(skb) + sizeof(struct iphdr)); if (unlikely(err)) return err; nh = ip_hdr(skb); - nwaddr = nla_type(a) == OVS_ACTION_ATTR_SET_NW_SRC ? &nh->saddr : &nh->daddr; - check = get_l4_checksum(skb); - if (likely(check)) - inet_proto_csum_replace4(check, skb, *nwaddr, new_nwaddr, 1); - csum_replace4(&nh->check, *nwaddr, new_nwaddr); + if (ipv4_key->ipv4_src != nh->saddr) + set_ip_addr(skb, nh, &nh->saddr, ipv4_key->ipv4_src); - skb_clear_rxhash(skb); + if (ipv4_key->ipv4_dst != nh->daddr) + set_ip_addr(skb, nh, &nh->daddr, ipv4_key->ipv4_dst); + + if (ipv4_key->ipv4_tos != nh->tos) + ipv4_change_dsfield(nh, 0, ipv4_key->ipv4_tos); - *nwaddr = new_nwaddr; + if (ipv4_key->ipv4_ttl != nh->ttl) + set_ip_ttl(skb, nh, ipv4_key->ipv4_ttl); return 0; } -static int set_nw_tos(struct sk_buff *skb, u8 nw_tos) +/* Must follow make_writable() since that can move the skb data. */ +static void set_tp_port(struct sk_buff *skb, __be16 *port, + __be16 new_port, __sum16 *check) { - struct iphdr *nh = ip_hdr(skb); - u8 old, new; - int err; + inet_proto_csum_replace2(check, skb, *port, new_port, 0); + *port = new_port; + skb_clear_rxhash(skb); +} - if (unlikely(!is_ip(skb))) - return 0; +static void set_udp_port(struct sk_buff *skb, __be16 *port, __be16 new_port) +{ + struct udphdr *uh = udp_hdr(skb); - err = make_writable(skb, skb_network_offset(skb) + - sizeof(struct iphdr)); + if (uh->check && get_ip_summed(skb) != OVS_CSUM_PARTIAL) { + set_tp_port(skb, port, new_port, &uh->check); + + if (!uh->check) + uh->check = CSUM_MANGLED_0; + } else { + *port = new_port; + skb_clear_rxhash(skb); + } +} + +static int set_udp(struct sk_buff *skb, const struct ovs_key_udp *udp_port_key) +{ + struct udphdr *uh; + int err; + + err = make_writable(skb, skb_transport_offset(skb) + + sizeof(struct udphdr)); if (unlikely(err)) return err; - /* Set the DSCP bits and preserve the ECN bits. */ - old = nh->tos; - new = nw_tos | (nh->tos & INET_ECN_MASK); - csum_replace4(&nh->check, (__force __be32)old, - (__force __be32)new); - nh->tos = new; + uh = udp_hdr(skb); + if (udp_port_key->udp_src != uh->source) + set_udp_port(skb, &uh->source, udp_port_key->udp_src); + + if (udp_port_key->udp_dst != uh->dest) + set_udp_port(skb, &uh->dest, udp_port_key->udp_dst); return 0; } -static int set_tp_port(struct sk_buff *skb, const struct nlattr *a) +static int set_tcp(struct sk_buff *skb, const struct ovs_key_tcp *tcp_port_key) { - struct udphdr *th; - __sum16 *check; - __be16 *port; + struct tcphdr *th; int err; - if (unlikely(!is_ip(skb))) - return 0; - err = make_writable(skb, skb_transport_offset(skb) + sizeof(struct tcphdr)); if (unlikely(err)) return err; - /* Must follow make_writable() since that can move the skb data. */ - check = get_l4_checksum(skb); - if (unlikely(!check)) - return 0; + th = tcp_hdr(skb); + if (tcp_port_key->tcp_src != th->source) + set_tp_port(skb, &th->source, tcp_port_key->tcp_src, &th->check); - /* - * Update port and checksum. - * - * This is OK because source and destination port numbers are at the - * same offsets in both UDP and TCP headers, and get_l4_checksum() only - * supports those protocols. - */ - th = udp_hdr(skb); - port = nla_type(a) == OVS_ACTION_ATTR_SET_TP_SRC ? &th->source : &th->dest; - inet_proto_csum_replace2(check, skb, *port, nla_get_be16(a), 0); - *port = nla_get_be16(a); - skb_clear_rxhash(skb); + if (tcp_port_key->tcp_dst != th->dest) + set_tp_port(skb, &th->dest, tcp_port_key->tcp_dst, &th->check); return 0; } @@ -235,13 +269,13 @@ static int do_output(struct datapath *dp, struct sk_buff *skb, int out_port) if (unlikely(!skb)) return -ENOMEM; - vport = rcu_dereference(dp->ports[out_port]); + vport = ovs_vport_rcu(dp, out_port); if (unlikely(!vport)) { kfree_skb(skb); return -ENODEV; } - vport_send(vport, skb); + ovs_vport_send(vport, skb); return 0; } @@ -270,7 +304,7 @@ static int output_userspace(struct datapath *dp, struct sk_buff *skb, } } - return dp_upcall(dp, skb, &upcall); + return ovs_dp_upcall(dp, skb, &upcall); } static int sample(struct datapath *dp, struct sk_buff *skb, @@ -298,6 +332,40 @@ static int sample(struct datapath *dp, struct sk_buff *skb, nla_len(acts_list), true); } +static int execute_set_action(struct sk_buff *skb, + const struct nlattr *nested_attr) +{ + int err = 0; + + switch (nla_type(nested_attr)) { + case OVS_KEY_ATTR_PRIORITY: + skb->priority = nla_get_u32(nested_attr); + break; + + case OVS_KEY_ATTR_TUN_ID: + OVS_CB(skb)->tun_id = nla_get_be64(nested_attr); + break; + + case OVS_KEY_ATTR_ETHERNET: + err = set_eth_addr(skb, nla_data(nested_attr)); + break; + + case OVS_KEY_ATTR_IPV4: + err = set_ipv4(skb, nla_data(nested_attr)); + break; + + case OVS_KEY_ATTR_TCP: + err = set_tcp(skb, nla_data(nested_attr)); + break; + + case OVS_KEY_ATTR_UDP: + err = set_udp(skb, nla_data(nested_attr)); + break; + } + + return err; +} + /* Execute a list of actions against 'skb'. */ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb, const struct nlattr *attr, int len, bool keep_skb) @@ -307,7 +375,6 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb, * then freeing the original skbuff is wasteful. So the following code * is slightly obscure just to avoid that. */ int prev_port = -1; - u32 priority = skb->priority; const struct nlattr *a; int rem; @@ -329,13 +396,9 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb, output_userspace(dp, skb, a); break; - case OVS_ACTION_ATTR_SET_TUNNEL: - OVS_CB(skb)->tun_id = nla_get_be64(a); - break; - case OVS_ACTION_ATTR_PUSH_VLAN: - err = push_vlan(skb, nla_get_be16(a)); - if (unlikely(err)) /* skb already freed */ + err = push_vlan(skb, nla_data(a)); + if (unlikely(err)) /* skb already freed. */ return err; break; @@ -343,45 +406,15 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb, err = pop_vlan(skb); break; - case OVS_ACTION_ATTR_SET_DL_SRC: - err = make_writable(skb, ETH_HLEN); - if (likely(!err)) - memcpy(eth_hdr(skb)->h_source, nla_data(a), ETH_ALEN); - break; - - case OVS_ACTION_ATTR_SET_DL_DST: - err = make_writable(skb, ETH_HLEN); - if (likely(!err)) - memcpy(eth_hdr(skb)->h_dest, nla_data(a), ETH_ALEN); - break; - - case OVS_ACTION_ATTR_SET_NW_SRC: - case OVS_ACTION_ATTR_SET_NW_DST: - err = set_nw_addr(skb, a); - break; - - case OVS_ACTION_ATTR_SET_NW_TOS: - err = set_nw_tos(skb, nla_get_u8(a)); - break; - - case OVS_ACTION_ATTR_SET_TP_SRC: - case OVS_ACTION_ATTR_SET_TP_DST: - err = set_tp_port(skb, a); - break; - - case OVS_ACTION_ATTR_SET_PRIORITY: - skb->priority = nla_get_u32(a); - break; - - case OVS_ACTION_ATTR_POP_PRIORITY: - skb->priority = priority; + case OVS_ACTION_ATTR_SET: + err = execute_set_action(skb, nla_data(a)); break; case OVS_ACTION_ATTR_SAMPLE: err = sample(dp, skb, a); break; - } + if (unlikely(err)) { kfree_skb(skb); return err; @@ -414,13 +447,13 @@ static int loop_suppress(struct datapath *dp, struct sw_flow_actions *actions) { if (net_ratelimit()) pr_warn("%s: flow looped %d times, dropping\n", - dp_name(dp), MAX_LOOPS); + ovs_dp_name(dp), MAX_LOOPS); actions->actions_len = 0; return -ELOOP; } /* Execute a list of actions against 'skb'. */ -int execute_actions(struct datapath *dp, struct sk_buff *skb) +int ovs_execute_actions(struct datapath *dp, struct sk_buff *skb) { struct sw_flow_actions *acts = rcu_dereference(OVS_CB(skb)->flow->sf_acts); struct loop_counter *loop;