From 33db4125ec745426c3483d6817d8f7ea5324cd05 Mon Sep 17 00:00:00 2001 From: Joe Stringer Date: Thu, 1 Oct 2015 15:00:37 -0700 Subject: openvswitch: Rename LABEL->LABELS Conntrack LABELS (plural) are exposed by conntrack; rename the OVS name for these to be consistent with conntrack. Fixes: c2ac667 "openvswitch: Allow matching on conntrack label" Signed-off-by: Joe Stringer Acked-by: Pravin B Shelar Signed-off-by: David S. Miller --- net/openvswitch/actions.c | 2 +- net/openvswitch/conntrack.c | 74 +++++++++++++++++++++--------------------- net/openvswitch/conntrack.h | 2 +- net/openvswitch/flow.h | 2 +- net/openvswitch/flow_netlink.c | 18 +++++----- 5 files changed, 49 insertions(+), 49 deletions(-) (limited to 'net/openvswitch') diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index 315f5330b6e5..e23a61cc3d5c 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -968,7 +968,7 @@ static int execute_masked_set_action(struct sk_buff *skb, case OVS_KEY_ATTR_CT_STATE: case OVS_KEY_ATTR_CT_ZONE: case OVS_KEY_ATTR_CT_MARK: - case OVS_KEY_ATTR_CT_LABEL: + case OVS_KEY_ATTR_CT_LABELS: err = -EINVAL; break; } diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c index 002a755fa07e..7d80acfb80d0 100644 --- a/net/openvswitch/conntrack.c +++ b/net/openvswitch/conntrack.c @@ -37,9 +37,9 @@ struct md_mark { }; /* Metadata label for masked write to conntrack label. */ -struct md_label { - struct ovs_key_ct_label value; - struct ovs_key_ct_label mask; +struct md_labels { + struct ovs_key_ct_labels value; + struct ovs_key_ct_labels mask; }; /* Conntrack action context for execution. */ @@ -50,7 +50,7 @@ struct ovs_conntrack_info { u32 flags; u16 family; struct md_mark mark; - struct md_label label; + struct md_labels labels; }; static u16 key_to_nfproto(const struct sw_flow_key *key) @@ -109,21 +109,21 @@ static u32 ovs_ct_get_mark(const struct nf_conn *ct) #endif } -static void ovs_ct_get_label(const struct nf_conn *ct, - struct ovs_key_ct_label *label) +static void ovs_ct_get_labels(const struct nf_conn *ct, + struct ovs_key_ct_labels *labels) { struct nf_conn_labels *cl = ct ? nf_ct_labels_find(ct) : NULL; if (cl) { size_t len = cl->words * sizeof(long); - if (len > OVS_CT_LABEL_LEN) - len = OVS_CT_LABEL_LEN; - else if (len < OVS_CT_LABEL_LEN) - memset(label, 0, OVS_CT_LABEL_LEN); - memcpy(label, cl->bits, len); + if (len > OVS_CT_LABELS_LEN) + len = OVS_CT_LABELS_LEN; + else if (len < OVS_CT_LABELS_LEN) + memset(labels, 0, OVS_CT_LABELS_LEN); + memcpy(labels, cl->bits, len); } else { - memset(label, 0, OVS_CT_LABEL_LEN); + memset(labels, 0, OVS_CT_LABELS_LEN); } } @@ -134,7 +134,7 @@ static void __ovs_ct_update_key(struct sw_flow_key *key, u8 state, key->ct.state = state; key->ct.zone = zone->id; key->ct.mark = ovs_ct_get_mark(ct); - ovs_ct_get_label(ct, &key->ct.label); + ovs_ct_get_labels(ct, &key->ct.labels); } /* Update 'key' based on skb->nfct. If 'post_ct' is true, then OVS has @@ -179,8 +179,8 @@ int ovs_ct_put_key(const struct sw_flow_key *key, struct sk_buff *skb) return -EMSGSIZE; if (IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS) && - nla_put(skb, OVS_KEY_ATTR_CT_LABEL, sizeof(key->ct.label), - &key->ct.label)) + nla_put(skb, OVS_KEY_ATTR_CT_LABELS, sizeof(key->ct.labels), + &key->ct.labels)) return -EMSGSIZE; return 0; @@ -213,9 +213,9 @@ static int ovs_ct_set_mark(struct sk_buff *skb, struct sw_flow_key *key, #endif } -static int ovs_ct_set_label(struct sk_buff *skb, struct sw_flow_key *key, - const struct ovs_key_ct_label *label, - const struct ovs_key_ct_label *mask) +static int ovs_ct_set_labels(struct sk_buff *skb, struct sw_flow_key *key, + const struct ovs_key_ct_labels *labels, + const struct ovs_key_ct_labels *mask) { enum ip_conntrack_info ctinfo; struct nf_conn_labels *cl; @@ -235,15 +235,15 @@ static int ovs_ct_set_label(struct sk_buff *skb, struct sw_flow_key *key, nf_ct_labels_ext_add(ct); cl = nf_ct_labels_find(ct); } - if (!cl || cl->words * sizeof(long) < OVS_CT_LABEL_LEN) + if (!cl || cl->words * sizeof(long) < OVS_CT_LABELS_LEN) return -ENOSPC; - err = nf_connlabels_replace(ct, (u32 *)label, (u32 *)mask, - OVS_CT_LABEL_LEN / sizeof(u32)); + err = nf_connlabels_replace(ct, (u32 *)labels, (u32 *)mask, + OVS_CT_LABELS_LEN / sizeof(u32)); if (err) return err; - ovs_ct_get_label(ct, &key->ct.label); + ovs_ct_get_labels(ct, &key->ct.labels); return 0; } @@ -465,12 +465,12 @@ static int ovs_ct_commit(struct net *net, struct sw_flow_key *key, return 0; } -static bool label_nonzero(const struct ovs_key_ct_label *label) +static bool labels_nonzero(const struct ovs_key_ct_labels *labels) { size_t i; - for (i = 0; i < sizeof(*label); i++) - if (label->ct_label[i]) + for (i = 0; i < sizeof(*labels); i++) + if (labels->ct_labels[i]) return true; return false; @@ -506,9 +506,9 @@ int ovs_ct_execute(struct net *net, struct sk_buff *skb, if (err) goto err; } - if (label_nonzero(&info->label.mask)) - err = ovs_ct_set_label(skb, key, &info->label.value, - &info->label.mask); + if (labels_nonzero(&info->labels.mask)) + err = ovs_ct_set_labels(skb, key, &info->labels.value, + &info->labels.mask); err: skb_push(skb, nh_ofs); return err; @@ -545,8 +545,8 @@ static const struct ovs_ct_len_tbl ovs_ct_attr_lens[OVS_CT_ATTR_MAX + 1] = { .maxlen = sizeof(u16) }, [OVS_CT_ATTR_MARK] = { .minlen = sizeof(struct md_mark), .maxlen = sizeof(struct md_mark) }, - [OVS_CT_ATTR_LABEL] = { .minlen = sizeof(struct md_label), - .maxlen = sizeof(struct md_label) }, + [OVS_CT_ATTR_LABELS] = { .minlen = sizeof(struct md_labels), + .maxlen = sizeof(struct md_labels) }, [OVS_CT_ATTR_HELPER] = { .minlen = 1, .maxlen = NF_CT_HELPER_NAME_LEN } }; @@ -593,10 +593,10 @@ static int parse_ct(const struct nlattr *attr, struct ovs_conntrack_info *info, } #endif #ifdef CONFIG_NF_CONNTRACK_LABELS - case OVS_CT_ATTR_LABEL: { - struct md_label *label = nla_data(a); + case OVS_CT_ATTR_LABELS: { + struct md_labels *labels = nla_data(a); - info->label = *label; + info->labels = *labels; break; } #endif @@ -633,7 +633,7 @@ bool ovs_ct_verify(struct net *net, enum ovs_key_attr attr) attr == OVS_KEY_ATTR_CT_MARK) return true; if (IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS) && - attr == OVS_KEY_ATTR_CT_LABEL) { + attr == OVS_KEY_ATTR_CT_LABELS) { struct ovs_net *ovs_net = net_generic(net, ovs_net_id); return ovs_net->xt_label; @@ -711,8 +711,8 @@ int ovs_ct_action_to_attr(const struct ovs_conntrack_info *ct_info, &ct_info->mark)) return -EMSGSIZE; if (IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS) && - nla_put(skb, OVS_CT_ATTR_LABEL, sizeof(ct_info->label), - &ct_info->label)) + nla_put(skb, OVS_CT_ATTR_LABELS, sizeof(ct_info->labels), + &ct_info->labels)) return -EMSGSIZE; if (ct_info->helper) { if (nla_put_string(skb, OVS_CT_ATTR_HELPER, @@ -737,7 +737,7 @@ void ovs_ct_free_action(const struct nlattr *a) void ovs_ct_init(struct net *net) { - unsigned int n_bits = sizeof(struct ovs_key_ct_label) * BITS_PER_BYTE; + unsigned int n_bits = sizeof(struct ovs_key_ct_labels) * BITS_PER_BYTE; struct ovs_net *ovs_net = net_generic(net, ovs_net_id); if (nf_connlabels_get(net, n_bits)) { diff --git a/net/openvswitch/conntrack.h b/net/openvswitch/conntrack.h index 43f5dd7a5577..6bd603c6a031 100644 --- a/net/openvswitch/conntrack.h +++ b/net/openvswitch/conntrack.h @@ -72,7 +72,7 @@ static inline void ovs_ct_fill_key(const struct sk_buff *skb, key->ct.state = 0; key->ct.zone = 0; key->ct.mark = 0; - memset(&key->ct.label, 0, sizeof(key->ct.label)); + memset(&key->ct.labels, 0, sizeof(key->ct.labels)); } static inline int ovs_ct_put_key(const struct sw_flow_key *key, diff --git a/net/openvswitch/flow.h b/net/openvswitch/flow.h index fe527d2dd4b7..8cfa15a08668 100644 --- a/net/openvswitch/flow.h +++ b/net/openvswitch/flow.h @@ -116,7 +116,7 @@ struct sw_flow_key { u16 zone; u32 mark; u8 state; - struct ovs_key_ct_label label; + struct ovs_key_ct_labels labels; } ct; } __aligned(BITS_PER_LONG/8); /* Ensure that we can do comparisons as longs. */ diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c index 5c030a4d7338..a60e3b7684bc 100644 --- a/net/openvswitch/flow_netlink.c +++ b/net/openvswitch/flow_netlink.c @@ -294,7 +294,7 @@ size_t ovs_key_attr_size(void) + nla_total_size(1) /* OVS_KEY_ATTR_CT_STATE */ + nla_total_size(2) /* OVS_KEY_ATTR_CT_ZONE */ + nla_total_size(4) /* OVS_KEY_ATTR_CT_MARK */ - + nla_total_size(16) /* OVS_KEY_ATTR_CT_LABEL */ + + nla_total_size(16) /* OVS_KEY_ATTR_CT_LABELS */ + nla_total_size(12) /* OVS_KEY_ATTR_ETHERNET */ + nla_total_size(2) /* OVS_KEY_ATTR_ETHERTYPE */ + nla_total_size(4) /* OVS_KEY_ATTR_VLAN */ @@ -352,7 +352,7 @@ static const struct ovs_len_tbl ovs_key_lens[OVS_KEY_ATTR_MAX + 1] = { [OVS_KEY_ATTR_CT_STATE] = { .len = sizeof(u8) }, [OVS_KEY_ATTR_CT_ZONE] = { .len = sizeof(u16) }, [OVS_KEY_ATTR_CT_MARK] = { .len = sizeof(u32) }, - [OVS_KEY_ATTR_CT_LABEL] = { .len = sizeof(struct ovs_key_ct_label) }, + [OVS_KEY_ATTR_CT_LABELS] = { .len = sizeof(struct ovs_key_ct_labels) }, }; static bool check_attr_len(unsigned int attr_len, unsigned int expected_len) @@ -833,14 +833,14 @@ static int metadata_from_nlattrs(struct net *net, struct sw_flow_match *match, SW_FLOW_KEY_PUT(match, ct.mark, mark, is_mask); *attrs &= ~(1ULL << OVS_KEY_ATTR_CT_MARK); } - if (*attrs & (1 << OVS_KEY_ATTR_CT_LABEL) && - ovs_ct_verify(net, OVS_KEY_ATTR_CT_LABEL)) { - const struct ovs_key_ct_label *cl; + if (*attrs & (1 << OVS_KEY_ATTR_CT_LABELS) && + ovs_ct_verify(net, OVS_KEY_ATTR_CT_LABELS)) { + const struct ovs_key_ct_labels *cl; - cl = nla_data(a[OVS_KEY_ATTR_CT_LABEL]); - SW_FLOW_KEY_MEMCPY(match, ct.label, cl->ct_label, + cl = nla_data(a[OVS_KEY_ATTR_CT_LABELS]); + SW_FLOW_KEY_MEMCPY(match, ct.labels, cl->ct_labels, sizeof(*cl), is_mask); - *attrs &= ~(1ULL << OVS_KEY_ATTR_CT_LABEL); + *attrs &= ~(1ULL << OVS_KEY_ATTR_CT_LABELS); } return 0; } @@ -1973,7 +1973,7 @@ static int validate_set(const struct nlattr *a, case OVS_KEY_ATTR_PRIORITY: case OVS_KEY_ATTR_SKB_MARK: case OVS_KEY_ATTR_CT_MARK: - case OVS_KEY_ATTR_CT_LABEL: + case OVS_KEY_ATTR_CT_LABELS: case OVS_KEY_ATTR_ETHERNET: break; -- cgit v1.2.1 From 598c12d0ba6de9060f04999746eb1e015774044b Mon Sep 17 00:00:00 2001 From: Konstantin Khlebnikov Date: Fri, 2 Oct 2015 13:18:22 +0300 Subject: ovs: do not allocate memory from offline numa node When openvswitch tries allocate memory from offline numa node 0: stats = kmem_cache_alloc_node(flow_stats_cache, GFP_KERNEL | __GFP_ZERO, 0) It catches VM_BUG_ON(nid < 0 || nid >= MAX_NUMNODES || !node_online(nid)) [ replaced with VM_WARN_ON(!node_online(nid)) recently ] in linux/gfp.h This patch disables numa affinity in this case. Signed-off-by: Konstantin Khlebnikov Acked-by: Pravin B Shelar Signed-off-by: David S. Miller --- net/openvswitch/flow_table.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net/openvswitch') diff --git a/net/openvswitch/flow_table.c b/net/openvswitch/flow_table.c index f2ea83ba4763..c7f74aab34b9 100644 --- a/net/openvswitch/flow_table.c +++ b/net/openvswitch/flow_table.c @@ -93,7 +93,8 @@ struct sw_flow *ovs_flow_alloc(void) /* Initialize the default stat node. */ stats = kmem_cache_alloc_node(flow_stats_cache, - GFP_KERNEL | __GFP_ZERO, 0); + GFP_KERNEL | __GFP_ZERO, + node_online(0) ? 0 : NUMA_NO_NODE); if (!stats) goto err; -- cgit v1.2.1 From 83ffe99f52b8f269b31b21524adcd13b165f7703 Mon Sep 17 00:00:00 2001 From: Pravin B Shelar Date: Fri, 2 Oct 2015 14:56:34 -0700 Subject: openvswitch: Fix ovs_vport_get_stats() Not every device has dev->tstats set. So when OVS tries to calculate vport stats it causes kernel panic. Following patch fixes it by using standard API to get net-device stats. ---8<--- Unable to handle kernel paging request at virtual address 766b4008 Internal error: Oops: 96000005 [#1] PREEMPT SMP Modules linked in: vport_vxlan vxlan ip6_udp_tunnel udp_tunnel tun bridge stp llc openvswitch ipv6 CPU: 7 PID: 1108 Comm: ovs-vswitchd Not tainted 4.3.0-rc3+ #82 PC is at ovs_vport_get_stats+0x150/0x1f8 [openvswitch] Call trace: [] ovs_vport_get_stats+0x150/0x1f8 [openvswitch] [] ovs_vport_cmd_fill_info+0x140/0x1e0 [openvswitch] [] ovs_vport_cmd_dump+0xbc/0x138 [openvswitch] [] netlink_dump+0xb8/0x258 [] __netlink_dump_start+0x120/0x178 [] genl_family_rcv_msg+0x2d4/0x308 [] genl_rcv_msg+0x88/0xc4 [] netlink_rcv_skb+0xd4/0x100 [] genl_rcv+0x30/0x48 [] netlink_unicast+0x154/0x200 [] netlink_sendmsg+0x308/0x364 [] sock_sendmsg+0x14/0x2c [] SyS_sendto+0xbc/0xf0 Code: aa1603e1 f94037a4 aa1303e2 aa1703e0 (f9400465) Reported-by: Tomasz Sawicki Fixes: 8c876639c98 ("openvswitch: Remove vport stats.") Signed-off-by: Pravin B Shelar Signed-off-by: David S. Miller --- net/openvswitch/vport.c | 42 +++++++++++++----------------------------- 1 file changed, 13 insertions(+), 29 deletions(-) (limited to 'net/openvswitch') diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c index dc81dc619aa2..fc5c0b9ccfe9 100644 --- a/net/openvswitch/vport.c +++ b/net/openvswitch/vport.c @@ -280,35 +280,19 @@ void ovs_vport_del(struct vport *vport) */ void ovs_vport_get_stats(struct vport *vport, struct ovs_vport_stats *stats) { - struct net_device *dev = vport->dev; - int i; - - memset(stats, 0, sizeof(*stats)); - stats->rx_errors = dev->stats.rx_errors; - stats->tx_errors = dev->stats.tx_errors; - stats->tx_dropped = dev->stats.tx_dropped; - stats->rx_dropped = dev->stats.rx_dropped; - - stats->rx_dropped += atomic_long_read(&dev->rx_dropped); - stats->tx_dropped += atomic_long_read(&dev->tx_dropped); - - for_each_possible_cpu(i) { - const struct pcpu_sw_netstats *percpu_stats; - struct pcpu_sw_netstats local_stats; - unsigned int start; - - percpu_stats = per_cpu_ptr(dev->tstats, i); - - do { - start = u64_stats_fetch_begin_irq(&percpu_stats->syncp); - local_stats = *percpu_stats; - } while (u64_stats_fetch_retry_irq(&percpu_stats->syncp, start)); - - stats->rx_bytes += local_stats.rx_bytes; - stats->rx_packets += local_stats.rx_packets; - stats->tx_bytes += local_stats.tx_bytes; - stats->tx_packets += local_stats.tx_packets; - } + const struct rtnl_link_stats64 *dev_stats; + struct rtnl_link_stats64 temp; + + dev_stats = dev_get_stats(vport->dev, &temp); + stats->rx_errors = dev_stats->rx_errors; + stats->tx_errors = dev_stats->tx_errors; + stats->tx_dropped = dev_stats->tx_dropped; + stats->rx_dropped = dev_stats->rx_dropped; + + stats->rx_bytes = dev_stats->rx_bytes; + stats->rx_packets = dev_stats->rx_packets; + stats->tx_bytes = dev_stats->tx_bytes; + stats->tx_packets = dev_stats->tx_packets; } /** -- cgit v1.2.1 From b8f2257069f179c7bdedc9501c1623070c4c37bb Mon Sep 17 00:00:00 2001 From: Joe Stringer Date: Tue, 6 Oct 2015 10:59:57 -0700 Subject: openvswitch: Fix skb leak in ovs_fragment() If ovs_fragment() was unable to fragment the skb due to an L2 header that exceeds the supported length, skbs would be leaked. Fix the bug. Fixes: 7f8a436eaa2c "openvswitch: Add conntrack action" Signed-off-by: Joe Stringer Acked-by: Pravin B Shelar Signed-off-by: David S. Miller --- net/openvswitch/actions.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'net/openvswitch') diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index e23a61cc3d5c..4cb93f92d6be 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -684,7 +684,7 @@ static void ovs_fragment(struct vport *vport, struct sk_buff *skb, u16 mru, { if (skb_network_offset(skb) > MAX_L2_LEN) { OVS_NLERR(1, "L2 header too long to fragment"); - return; + goto err; } if (ethertype == htons(ETH_P_IP)) { @@ -708,8 +708,7 @@ static void ovs_fragment(struct vport *vport, struct sk_buff *skb, u16 mru, struct rt6_info ovs_rt; if (!v6ops) { - kfree_skb(skb); - return; + goto err; } prepare_frag(vport, skb); @@ -728,8 +727,12 @@ static void ovs_fragment(struct vport *vport, struct sk_buff *skb, u16 mru, WARN_ONCE(1, "Failed fragment ->%s: eth=%04x, MRU=%d, MTU=%d.", ovs_vport_name(vport), ntohs(ethertype), mru, vport->dev->mtu); - kfree_skb(skb); + goto err; } + + return; +err: + kfree_skb(skb); } static void do_output(struct datapath *dp, struct sk_buff *skb, int out_port, -- cgit v1.2.1 From ec0d043d05e6e3c0c2fac5de922c800c027c6386 Mon Sep 17 00:00:00 2001 From: Joe Stringer Date: Tue, 6 Oct 2015 10:59:58 -0700 Subject: openvswitch: Ensure flow is valid before executing ct The ct action uses parts of the flow key, so we need to ensure that it is valid before executing that action. Fixes: 7f8a436eaa2c "openvswitch: Add conntrack action" Signed-off-by: Joe Stringer Acked-by: Pravin B Shelar Signed-off-by: David S. Miller --- net/openvswitch/actions.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'net/openvswitch') diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index 4cb93f92d6be..c6a39bf2c3b9 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -1102,6 +1102,12 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb, break; case OVS_ACTION_ATTR_CT: + if (!is_flow_key_valid(key)) { + err = ovs_flow_key_update(skb, key); + if (err) + return err; + } + err = ovs_ct_execute(ovs_dp_get_net(dp), skb, key, nla_data(a)); -- cgit v1.2.1 From 6f225952461b5e9b5520d0dc6e2ff0af57874fbb Mon Sep 17 00:00:00 2001 From: Joe Stringer Date: Tue, 6 Oct 2015 10:59:59 -0700 Subject: openvswitch: Reject ct_state unsupported bits Previously, if userspace specified ct_state bits in the flow key which are currently undefined (and therefore unsupported), then they would be ignored. This could cause unexpected behaviour in future if userspace is extended to support additional bits but attempts to communicate with the current version of the kernel. This patch rectifies the situation by rejecting such ct_state bits. Fixes: 7f8a436eaa2c "openvswitch: Add conntrack action" Signed-off-by: Joe Stringer Acked-by: Pravin B Shelar Signed-off-by: David S. Miller --- net/openvswitch/conntrack.h | 12 ++++++++++++ net/openvswitch/flow_netlink.c | 6 ++++++ 2 files changed, 18 insertions(+) (limited to 'net/openvswitch') diff --git a/net/openvswitch/conntrack.h b/net/openvswitch/conntrack.h index 6bd603c6a031..d6eca8394254 100644 --- a/net/openvswitch/conntrack.h +++ b/net/openvswitch/conntrack.h @@ -34,6 +34,13 @@ int ovs_ct_execute(struct net *, struct sk_buff *, struct sw_flow_key *, void ovs_ct_fill_key(const struct sk_buff *skb, struct sw_flow_key *key); int ovs_ct_put_key(const struct sw_flow_key *key, struct sk_buff *skb); void ovs_ct_free_action(const struct nlattr *a); + +static inline bool ovs_ct_state_supported(u8 state) +{ + return !(state & ~(OVS_CS_F_NEW | OVS_CS_F_ESTABLISHED | + OVS_CS_F_RELATED | OVS_CS_F_REPLY_DIR | + OVS_CS_F_INVALID | OVS_CS_F_TRACKED)); +} #else #include @@ -46,6 +53,11 @@ static inline bool ovs_ct_verify(struct net *net, int attr) return false; } +static inline bool ovs_ct_state_supported(u8 state) +{ + return false; +} + static inline int ovs_ct_copy_action(struct net *net, const struct nlattr *nla, const struct sw_flow_key *key, struct sw_flow_actions **acts, bool log) diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c index a60e3b7684bc..d47b5c5c640e 100644 --- a/net/openvswitch/flow_netlink.c +++ b/net/openvswitch/flow_netlink.c @@ -816,6 +816,12 @@ static int metadata_from_nlattrs(struct net *net, struct sw_flow_match *match, ovs_ct_verify(net, OVS_KEY_ATTR_CT_STATE)) { u8 ct_state = nla_get_u8(a[OVS_KEY_ATTR_CT_STATE]); + if (!is_mask && !ovs_ct_state_supported(ct_state)) { + OVS_NLERR(log, "ct_state flags %02x unsupported", + ct_state); + return -EINVAL; + } + SW_FLOW_KEY_PUT(match, ct.state, ct_state, is_mask); *attrs &= ~(1ULL << OVS_KEY_ATTR_CT_STATE); } -- cgit v1.2.1 From fbccce5965a58d56aaed9e9acd1bec75d8a66e87 Mon Sep 17 00:00:00 2001 From: Joe Stringer Date: Tue, 6 Oct 2015 11:00:00 -0700 Subject: openvswitch: Extend ct_state match field to 32 bits The ct_state field was initially added as an 8-bit field, however six of the bits are already being used and use cases are already starting to appear that may push the limits of this field. This patch extends the field to 32 bits while retaining the internal representation of 8 bits. This should cover forward compatibility of the ABI for the foreseeable future. This patch also reorders the OVS_CS_F_* bits to be sequential. Suggested-by: Jarno Rajahalme Signed-off-by: Joe Stringer Acked-by: Pravin B Shelar Signed-off-by: David S. Miller --- net/openvswitch/conntrack.c | 2 +- net/openvswitch/conntrack.h | 4 ++-- net/openvswitch/flow_netlink.c | 8 ++++---- 3 files changed, 7 insertions(+), 7 deletions(-) (limited to 'net/openvswitch') diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c index 7d80acfb80d0..466d5576fe3f 100644 --- a/net/openvswitch/conntrack.c +++ b/net/openvswitch/conntrack.c @@ -167,7 +167,7 @@ void ovs_ct_fill_key(const struct sk_buff *skb, struct sw_flow_key *key) int ovs_ct_put_key(const struct sw_flow_key *key, struct sk_buff *skb) { - if (nla_put_u8(skb, OVS_KEY_ATTR_CT_STATE, key->ct.state)) + if (nla_put_u32(skb, OVS_KEY_ATTR_CT_STATE, key->ct.state)) return -EMSGSIZE; if (IS_ENABLED(CONFIG_NF_CONNTRACK_ZONES) && diff --git a/net/openvswitch/conntrack.h b/net/openvswitch/conntrack.h index d6eca8394254..da8714942c95 100644 --- a/net/openvswitch/conntrack.h +++ b/net/openvswitch/conntrack.h @@ -35,7 +35,7 @@ void ovs_ct_fill_key(const struct sk_buff *skb, struct sw_flow_key *key); int ovs_ct_put_key(const struct sw_flow_key *key, struct sk_buff *skb); void ovs_ct_free_action(const struct nlattr *a); -static inline bool ovs_ct_state_supported(u8 state) +static inline bool ovs_ct_state_supported(u32 state) { return !(state & ~(OVS_CS_F_NEW | OVS_CS_F_ESTABLISHED | OVS_CS_F_RELATED | OVS_CS_F_REPLY_DIR | @@ -53,7 +53,7 @@ static inline bool ovs_ct_verify(struct net *net, int attr) return false; } -static inline bool ovs_ct_state_supported(u8 state) +static inline bool ovs_ct_state_supported(u32 state) { return false; } diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c index d47b5c5c640e..171a691f1c32 100644 --- a/net/openvswitch/flow_netlink.c +++ b/net/openvswitch/flow_netlink.c @@ -291,7 +291,7 @@ size_t ovs_key_attr_size(void) + nla_total_size(4) /* OVS_KEY_ATTR_SKB_MARK */ + nla_total_size(4) /* OVS_KEY_ATTR_DP_HASH */ + nla_total_size(4) /* OVS_KEY_ATTR_RECIRC_ID */ - + nla_total_size(1) /* OVS_KEY_ATTR_CT_STATE */ + + nla_total_size(4) /* OVS_KEY_ATTR_CT_STATE */ + nla_total_size(2) /* OVS_KEY_ATTR_CT_ZONE */ + nla_total_size(4) /* OVS_KEY_ATTR_CT_MARK */ + nla_total_size(16) /* OVS_KEY_ATTR_CT_LABELS */ @@ -349,7 +349,7 @@ static const struct ovs_len_tbl ovs_key_lens[OVS_KEY_ATTR_MAX + 1] = { [OVS_KEY_ATTR_TUNNEL] = { .len = OVS_ATTR_NESTED, .next = ovs_tunnel_key_lens, }, [OVS_KEY_ATTR_MPLS] = { .len = sizeof(struct ovs_key_mpls) }, - [OVS_KEY_ATTR_CT_STATE] = { .len = sizeof(u8) }, + [OVS_KEY_ATTR_CT_STATE] = { .len = sizeof(u32) }, [OVS_KEY_ATTR_CT_ZONE] = { .len = sizeof(u16) }, [OVS_KEY_ATTR_CT_MARK] = { .len = sizeof(u32) }, [OVS_KEY_ATTR_CT_LABELS] = { .len = sizeof(struct ovs_key_ct_labels) }, @@ -814,10 +814,10 @@ static int metadata_from_nlattrs(struct net *net, struct sw_flow_match *match, if (*attrs & (1 << OVS_KEY_ATTR_CT_STATE) && ovs_ct_verify(net, OVS_KEY_ATTR_CT_STATE)) { - u8 ct_state = nla_get_u8(a[OVS_KEY_ATTR_CT_STATE]); + u32 ct_state = nla_get_u32(a[OVS_KEY_ATTR_CT_STATE]); if (!is_mask && !ovs_ct_state_supported(ct_state)) { - OVS_NLERR(log, "ct_state flags %02x unsupported", + OVS_NLERR(log, "ct_state flags %08x unsupported", ct_state); return -EINVAL; } -- cgit v1.2.1 From ab38a7b5a4493a3658d891a8e91f9ffcb3d2defb Mon Sep 17 00:00:00 2001 From: Joe Stringer Date: Tue, 6 Oct 2015 11:00:01 -0700 Subject: openvswitch: Change CT_ATTR_FLAGS to CT_ATTR_COMMIT Previously, the CT_ATTR_FLAGS attribute, when nested under the OVS_ACTION_ATTR_CT, encoded a 32-bit bitmask of flags that modify the semantics of the ct action. It's more extensible to just represent each flag as a nested attribute, and this requires no additional error checking to reject flags that aren't currently supported. Suggested-by: Ben Pfaff Signed-off-by: Joe Stringer Acked-by: Pravin B Shelar Signed-off-by: David S. Miller --- net/openvswitch/conntrack.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-) (limited to 'net/openvswitch') diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c index 466d5576fe3f..80bf702715bb 100644 --- a/net/openvswitch/conntrack.c +++ b/net/openvswitch/conntrack.c @@ -47,7 +47,7 @@ struct ovs_conntrack_info { struct nf_conntrack_helper *helper; struct nf_conntrack_zone zone; struct nf_conn *ct; - u32 flags; + u8 commit : 1; u16 family; struct md_mark mark; struct md_labels labels; @@ -493,7 +493,7 @@ int ovs_ct_execute(struct net *net, struct sk_buff *skb, return err; } - if (info->flags & OVS_CT_F_COMMIT) + if (info->commit) err = ovs_ct_commit(net, key, info, skb); else err = ovs_ct_lookup(net, key, info, skb); @@ -539,8 +539,7 @@ static int ovs_ct_add_helper(struct ovs_conntrack_info *info, const char *name, } static const struct ovs_ct_len_tbl ovs_ct_attr_lens[OVS_CT_ATTR_MAX + 1] = { - [OVS_CT_ATTR_FLAGS] = { .minlen = sizeof(u32), - .maxlen = sizeof(u32) }, + [OVS_CT_ATTR_COMMIT] = { .minlen = 0, .maxlen = 0 }, [OVS_CT_ATTR_ZONE] = { .minlen = sizeof(u16), .maxlen = sizeof(u16) }, [OVS_CT_ATTR_MARK] = { .minlen = sizeof(struct md_mark), @@ -576,8 +575,8 @@ static int parse_ct(const struct nlattr *attr, struct ovs_conntrack_info *info, } switch (type) { - case OVS_CT_ATTR_FLAGS: - info->flags = nla_get_u32(a); + case OVS_CT_ATTR_COMMIT: + info->commit = true; break; #ifdef CONFIG_NF_CONNTRACK_ZONES case OVS_CT_ATTR_ZONE: @@ -701,7 +700,7 @@ int ovs_ct_action_to_attr(const struct ovs_conntrack_info *ct_info, if (!start) return -EMSGSIZE; - if (nla_put_u32(skb, OVS_CT_ATTR_FLAGS, ct_info->flags)) + if (ct_info->commit && nla_put_flag(skb, OVS_CT_ATTR_COMMIT)) return -EMSGSIZE; if (IS_ENABLED(CONFIG_NF_CONNTRACK_ZONES) && nla_put_u16(skb, OVS_CT_ATTR_ZONE, ct_info->zone.id)) -- cgit v1.2.1 From 740dbc289155fdeed32438396370e70b684cd45e Mon Sep 17 00:00:00 2001 From: Joe Stringer Date: Fri, 16 Oct 2015 11:08:18 -0700 Subject: openvswitch: Scrub skb between namespaces If OVS receives a packet from another namespace, then the packet should be scrubbed. However, people have already begun to rely on the behaviour that skb->mark is preserved across namespaces, so retain this one field. This is mainly to address information leakage between namespaces when using OVS internal ports, but by placing it in ovs_vport_receive() it is more generally applicable, meaning it should not be overlooked if other port types are allowed to be moved into namespaces in future. Signed-off-by: Joe Stringer Acked-by: Pravin B Shelar Acked-by: Thomas Graf Signed-off-by: David S. Miller --- net/openvswitch/vport.c | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'net/openvswitch') diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c index fc5c0b9ccfe9..12a36ac21eda 100644 --- a/net/openvswitch/vport.c +++ b/net/openvswitch/vport.c @@ -444,6 +444,15 @@ int ovs_vport_receive(struct vport *vport, struct sk_buff *skb, OVS_CB(skb)->input_vport = vport; OVS_CB(skb)->mru = 0; + if (unlikely(dev_net(skb->dev) != ovs_dp_get_net(vport->dp))) { + u32 mark; + + mark = skb->mark; + skb_scrub_packet(skb, true); + skb->mark = mark; + tun_info = NULL; + } + /* Extract flow from 'skb' into 'key'. */ error = ovs_flow_key_extract(tun_info, skb, &key); if (unlikely(error)) { -- cgit v1.2.1