From 7016e0627171878810798a842a416dddee4e3329 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Wed, 13 Sep 2017 13:58:15 -0700 Subject: net: Convert int functions to bool Global function ipv6_rcv_saddr_equal and static functions ipv6_rcv_saddr_equal and ipv4_rcv_saddr_equal currently return int. bool is slightly more descriptive for these functions so change their return type from int to bool. Signed-off-by: Joe Perches Signed-off-by: David S. Miller --- include/net/addrconf.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index f44ff2476758..87981cd63180 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -94,8 +94,8 @@ int __ipv6_get_lladdr(struct inet6_dev *idev, struct in6_addr *addr, u32 banned_flags); int ipv6_get_lladdr(struct net_device *dev, struct in6_addr *addr, u32 banned_flags); -int inet_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2, - bool match_wildcard); +bool inet_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2, + bool match_wildcard); void addrconf_join_solict(struct net_device *dev, const struct in6_addr *addr); void addrconf_leave_solict(struct inet6_dev *idev, const struct in6_addr *addr); -- cgit v1.2.3 From bffa72cf7f9df842f0016ba03586039296b4caaf Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 19 Sep 2017 05:14:24 -0700 Subject: net: sk_buff rbnode reorg skb->rbnode shares space with skb->next, skb->prev and skb->tstamp Current uses (TCP receive ofo queue and netem) need to save/restore tstamp, while skb->dev is either NULL (TCP) or a constant for a given queue (netem). Since we plan using an RB tree for TCP retransmit queue to speedup SACK processing with large BDP, this patch exchanges skb->dev and skb->tstamp. This saves some overhead in both TCP and netem. v2: removes the swtstamp field from struct tcp_skb_cb Signed-off-by: Eric Dumazet Cc: Soheil Hassas Yeganeh Cc: Wei Wang Cc: Willem de Bruijn Acked-by: Soheil Hassas Yeganeh Signed-off-by: David S. Miller --- include/linux/skbuff.h | 16 ++++++++-------- include/net/tcp.h | 6 ------ net/ipv4/tcp_input.c | 27 +++++---------------------- net/sched/sch_netem.c | 7 ++++--- 4 files changed, 17 insertions(+), 39 deletions(-) (limited to 'include/net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 72299ef00061..492828801acb 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -661,8 +661,12 @@ struct sk_buff { struct sk_buff *prev; union { - ktime_t tstamp; - u64 skb_mstamp; + struct net_device *dev; + /* Some protocols might use this space to store information, + * while device pointer would be NULL. + * UDP receive path is one user. + */ + unsigned long dev_scratch; }; }; struct rb_node rbnode; /* used in netem & tcp stack */ @@ -670,12 +674,8 @@ struct sk_buff { struct sock *sk; union { - struct net_device *dev; - /* Some protocols might use this space to store information, - * while device pointer would be NULL. - * UDP receive path is one user. - */ - unsigned long dev_scratch; + ktime_t tstamp; + u64 skb_mstamp; }; /* * This is the control buffer. It is free to use for every diff --git a/include/net/tcp.h b/include/net/tcp.h index b510f284427a..49a8a46466f3 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -797,12 +797,6 @@ struct tcp_skb_cb { u16 tcp_gso_segs; u16 tcp_gso_size; }; - - /* Used to stash the receive timestamp while this skb is in the - * out of order queue, as skb->tstamp is overwritten by the - * rbnode. - */ - ktime_t swtstamp; }; __u8 tcp_flags; /* TCP header flags. (tcp[13]) */ diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index bddf724f5c02..db9bb46b5776 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -4266,11 +4266,6 @@ static void tcp_sack_remove(struct tcp_sock *tp) tp->rx_opt.num_sacks = num_sacks; } -enum tcp_queue { - OOO_QUEUE, - RCV_QUEUE, -}; - /** * tcp_try_coalesce - try to merge skb to prior one * @sk: socket @@ -4286,7 +4281,6 @@ enum tcp_queue { * Returns true if caller should free @from instead of queueing it */ static bool tcp_try_coalesce(struct sock *sk, - enum tcp_queue dest, struct sk_buff *to, struct sk_buff *from, bool *fragstolen) @@ -4311,10 +4305,7 @@ static bool tcp_try_coalesce(struct sock *sk, if (TCP_SKB_CB(from)->has_rxtstamp) { TCP_SKB_CB(to)->has_rxtstamp = true; - if (dest == OOO_QUEUE) - TCP_SKB_CB(to)->swtstamp = TCP_SKB_CB(from)->swtstamp; - else - to->tstamp = from->tstamp; + to->tstamp = from->tstamp; } return true; @@ -4351,9 +4342,6 @@ static void tcp_ofo_queue(struct sock *sk) } p = rb_next(p); rb_erase(&skb->rbnode, &tp->out_of_order_queue); - /* Replace tstamp which was stomped by rbnode */ - if (TCP_SKB_CB(skb)->has_rxtstamp) - skb->tstamp = TCP_SKB_CB(skb)->swtstamp; if (unlikely(!after(TCP_SKB_CB(skb)->end_seq, tp->rcv_nxt))) { SOCK_DEBUG(sk, "ofo packet was already received\n"); @@ -4365,8 +4353,7 @@ static void tcp_ofo_queue(struct sock *sk) TCP_SKB_CB(skb)->end_seq); tail = skb_peek_tail(&sk->sk_receive_queue); - eaten = tail && tcp_try_coalesce(sk, RCV_QUEUE, - tail, skb, &fragstolen); + eaten = tail && tcp_try_coalesce(sk, tail, skb, &fragstolen); tcp_rcv_nxt_update(tp, TCP_SKB_CB(skb)->end_seq); fin = TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN; if (!eaten) @@ -4420,10 +4407,6 @@ static void tcp_data_queue_ofo(struct sock *sk, struct sk_buff *skb) return; } - /* Stash tstamp to avoid being stomped on by rbnode */ - if (TCP_SKB_CB(skb)->has_rxtstamp) - TCP_SKB_CB(skb)->swtstamp = skb->tstamp; - /* Disable header prediction. */ tp->pred_flags = 0; inet_csk_schedule_ack(sk); @@ -4451,7 +4434,7 @@ static void tcp_data_queue_ofo(struct sock *sk, struct sk_buff *skb) /* In the typical case, we are adding an skb to the end of the list. * Use of ooo_last_skb avoids the O(Log(N)) rbtree lookup. */ - if (tcp_try_coalesce(sk, OOO_QUEUE, tp->ooo_last_skb, + if (tcp_try_coalesce(sk, tp->ooo_last_skb, skb, &fragstolen)) { coalesce_done: tcp_grow_window(sk, skb); @@ -4502,7 +4485,7 @@ coalesce_done: __kfree_skb(skb1); goto merge_right; } - } else if (tcp_try_coalesce(sk, OOO_QUEUE, skb1, + } else if (tcp_try_coalesce(sk, skb1, skb, &fragstolen)) { goto coalesce_done; } @@ -4554,7 +4537,7 @@ static int __must_check tcp_queue_rcv(struct sock *sk, struct sk_buff *skb, int __skb_pull(skb, hdrlen); eaten = (tail && - tcp_try_coalesce(sk, RCV_QUEUE, tail, + tcp_try_coalesce(sk, tail, skb, fragstolen)) ? 1 : 0; tcp_rcv_nxt_update(tcp_sk(sk), TCP_SKB_CB(skb)->end_seq); if (!eaten) { diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index b1266e75ca43..063a4bdb9ee6 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -146,7 +146,6 @@ struct netem_sched_data { */ struct netem_skb_cb { psched_time_t time_to_send; - ktime_t tstamp_save; }; @@ -561,7 +560,6 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch, } cb->time_to_send = now + delay; - cb->tstamp_save = skb->tstamp; ++q->counter; tfifo_enqueue(skb, sch); } else { @@ -629,7 +627,10 @@ deliver: qdisc_qstats_backlog_dec(sch, skb); skb->next = NULL; skb->prev = NULL; - skb->tstamp = netem_skb_cb(skb)->tstamp_save; + /* skb->dev shares skb->rbnode area, + * we need to restore its value. + */ + skb->dev = qdisc_dev(sch); #ifdef CONFIG_NET_CLS_ACT /* -- cgit v1.2.3 From f5619866592c65adc087364cc1a3ba709201ea26 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Tue, 19 Sep 2017 11:56:57 -0400 Subject: net: dsa: remove copy of master ethtool_ops There is no need to store a copy of the master ethtool ops, storing the original pointer in DSA and the new one in the master netdev itself is enough. In the meantime, set orig_ethtool_ops to NULL when restoring the master ethtool ops and check the presence of the master original ethtool ops as well as its needed functions before calling them. Signed-off-by: Vivien Didelot Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- include/net/dsa.h | 1 - net/dsa/dsa.c | 8 ++++---- net/dsa/slave.c | 19 +++++++++++-------- 3 files changed, 15 insertions(+), 13 deletions(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index dd44d6ce1097..8dee216a5a9b 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -188,7 +188,6 @@ struct dsa_port { /* * Original copy of the master netdev ethtool_ops */ - struct ethtool_ops ethtool_ops; const struct ethtool_ops *orig_ethtool_ops; }; diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 03c58b0eb082..abadf7b49236 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -124,11 +124,10 @@ int dsa_cpu_port_ethtool_setup(struct dsa_port *cpu_dp) if (!cpu_ops) return -ENOMEM; - memcpy(&cpu_dp->ethtool_ops, master->ethtool_ops, - sizeof(struct ethtool_ops)); cpu_dp->orig_ethtool_ops = master->ethtool_ops; - memcpy(cpu_ops, &cpu_dp->ethtool_ops, - sizeof(struct ethtool_ops)); + if (cpu_dp->orig_ethtool_ops) + memcpy(cpu_ops, cpu_dp->orig_ethtool_ops, sizeof(*cpu_ops)); + dsa_cpu_port_ethtool_init(cpu_ops); master->ethtool_ops = cpu_ops; @@ -138,6 +137,7 @@ int dsa_cpu_port_ethtool_setup(struct dsa_port *cpu_dp) void dsa_cpu_port_ethtool_restore(struct dsa_port *cpu_dp) { cpu_dp->netdev->ethtool_ops = cpu_dp->orig_ethtool_ops; + cpu_dp->orig_ethtool_ops = NULL; } void dsa_cpu_dsa_destroy(struct dsa_port *port) diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 2afa99506f8b..2ff4f907d137 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -574,12 +574,13 @@ static void dsa_cpu_port_get_ethtool_stats(struct net_device *dev, struct dsa_switch_tree *dst = dev->dsa_ptr; struct dsa_port *cpu_dp = dsa_get_cpu_port(dst); struct dsa_switch *ds = cpu_dp->ds; + const struct ethtool_ops *ops = cpu_dp->orig_ethtool_ops; s8 cpu_port = cpu_dp->index; int count = 0; - if (cpu_dp->ethtool_ops.get_sset_count) { - count = cpu_dp->ethtool_ops.get_sset_count(dev, ETH_SS_STATS); - cpu_dp->ethtool_ops.get_ethtool_stats(dev, stats, data); + if (ops && ops->get_sset_count && ops->get_ethtool_stats) { + count = ops->get_sset_count(dev, ETH_SS_STATS); + ops->get_ethtool_stats(dev, stats, data); } if (ds->ops->get_ethtool_stats) @@ -591,10 +592,11 @@ static int dsa_cpu_port_get_sset_count(struct net_device *dev, int sset) struct dsa_switch_tree *dst = dev->dsa_ptr; struct dsa_port *cpu_dp = dsa_get_cpu_port(dst); struct dsa_switch *ds = cpu_dp->ds; + const struct ethtool_ops *ops = cpu_dp->orig_ethtool_ops; int count = 0; - if (cpu_dp->ethtool_ops.get_sset_count) - count += cpu_dp->ethtool_ops.get_sset_count(dev, sset); + if (ops && ops->get_sset_count) + count += ops->get_sset_count(dev, sset); if (sset == ETH_SS_STATS && ds->ops->get_sset_count) count += ds->ops->get_sset_count(ds); @@ -608,6 +610,7 @@ static void dsa_cpu_port_get_strings(struct net_device *dev, struct dsa_switch_tree *dst = dev->dsa_ptr; struct dsa_port *cpu_dp = dsa_get_cpu_port(dst); struct dsa_switch *ds = cpu_dp->ds; + const struct ethtool_ops *ops = cpu_dp->orig_ethtool_ops; s8 cpu_port = cpu_dp->index; int len = ETH_GSTRING_LEN; int mcount = 0, count; @@ -619,9 +622,9 @@ static void dsa_cpu_port_get_strings(struct net_device *dev, /* We do not want to be NULL-terminated, since this is a prefix */ pfx[sizeof(pfx) - 1] = '_'; - if (cpu_dp->ethtool_ops.get_sset_count) { - mcount = cpu_dp->ethtool_ops.get_sset_count(dev, ETH_SS_STATS); - cpu_dp->ethtool_ops.get_strings(dev, stringset, data); + if (ops && ops->get_sset_count && ops->get_strings) { + mcount = ops->get_sset_count(dev, ETH_SS_STATS); + ops->get_strings(dev, stringset, data); } if (stringset == ETH_SS_STATS && ds->ops->get_strings) { -- cgit v1.2.3 From 752fbcc33405d6f8249465e4b2c4e420091bb825 Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Tue, 19 Sep 2017 13:15:42 -0700 Subject: net_sched: no need to free qdisc in RCU callback gen estimator has been rewritten in commit 1c0d32fde5bd ("net_sched: gen_estimator: complete rewrite of rate estimators"), the caller no longer needs to wait for a grace period. So this patch gets rid of it. Cc: Jamal Hadi Salim Cc: Eric Dumazet Signed-off-by: Cong Wang Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/sch_generic.h | 1 - net/sched/sch_generic.c | 10 ++-------- 2 files changed, 2 insertions(+), 9 deletions(-) (limited to 'include/net') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 135f5a2dd931..684d8ed27eaa 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -93,7 +93,6 @@ struct Qdisc { unsigned long state; struct Qdisc *next_sched; struct sk_buff *skb_bad_txq; - struct rcu_head rcu_head; int padded; refcount_t refcnt; diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 92237e75dbbc..1fb0c754b7fd 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -688,10 +688,8 @@ void qdisc_reset(struct Qdisc *qdisc) } EXPORT_SYMBOL(qdisc_reset); -static void qdisc_rcu_free(struct rcu_head *head) +static void qdisc_free(struct Qdisc *qdisc) { - struct Qdisc *qdisc = container_of(head, struct Qdisc, rcu_head); - if (qdisc_is_percpu_stats(qdisc)) { free_percpu(qdisc->cpu_bstats); free_percpu(qdisc->cpu_qstats); @@ -724,11 +722,7 @@ void qdisc_destroy(struct Qdisc *qdisc) kfree_skb_list(qdisc->gso_skb); kfree_skb(qdisc->skb_bad_txq); - /* - * gen_estimator est_timer() might access qdisc->q.lock, - * wait a RCU grace period before freeing qdisc. - */ - call_rcu(&qdisc->rcu_head, qdisc_rcu_free); + qdisc_free(qdisc); } EXPORT_SYMBOL(qdisc_destroy); -- cgit v1.2.3 From a90c9347e90ed1e9323d71402ed18023bc910cd8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 19 Sep 2017 16:27:06 -0700 Subject: ipv6: addrlabel: per netns list Having a global list of labels do not scale to thousands of netns in the cloud era. This causes quadratic behavior on netns creation and deletion. This is time having a per netns list of ~10 labels. Tested: $ time perf record (for f in `seq 1 3000` ; do ip netns add tast$f; done) [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 3.637 MB perf.data (~158898 samples) ] real 0m20.837s # instead of 0m24.227s user 0m0.328s sys 0m20.338s # instead of 0m23.753s 16.17% ip [kernel.kallsyms] [k] netlink_broadcast_filtered 12.30% ip [kernel.kallsyms] [k] netlink_has_listeners 6.76% ip [kernel.kallsyms] [k] _raw_spin_lock_irqsave 5.78% ip [kernel.kallsyms] [k] memset_erms 5.77% ip [kernel.kallsyms] [k] kobject_uevent_env 5.18% ip [kernel.kallsyms] [k] refcount_sub_and_test 4.96% ip [kernel.kallsyms] [k] _raw_read_lock 3.82% ip [kernel.kallsyms] [k] refcount_inc_not_zero 3.33% ip [kernel.kallsyms] [k] _raw_spin_unlock_irqrestore 2.11% ip [kernel.kallsyms] [k] unmap_page_range 1.77% ip [kernel.kallsyms] [k] __wake_up 1.69% ip [kernel.kallsyms] [k] strlen 1.17% ip [kernel.kallsyms] [k] __wake_up_common 1.09% ip [kernel.kallsyms] [k] insert_header 1.04% ip [kernel.kallsyms] [k] page_remove_rmap 1.01% ip [kernel.kallsyms] [k] consume_skb 0.98% ip [kernel.kallsyms] [k] netlink_trim 0.51% ip [kernel.kallsyms] [k] kernfs_link_sibling 0.51% ip [kernel.kallsyms] [k] filemap_map_pages 0.46% ip [kernel.kallsyms] [k] memcpy_erms Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv6.h | 5 +++ net/ipv6/addrlabel.c | 81 ++++++++++++++++++------------------------------ 2 files changed, 35 insertions(+), 51 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index 2544f9760a42..2ea1ed341ef8 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -89,6 +89,11 @@ struct netns_ipv6 { atomic_t fib6_sernum; struct seg6_pernet_data *seg6_data; struct fib_notifier_ops *notifier_ops; + struct { + struct hlist_head head; + spinlock_t lock; + u32 seq; + } ip6addrlbl_table; }; #if IS_ENABLED(CONFIG_NF_DEFRAG_IPV6) diff --git a/net/ipv6/addrlabel.c b/net/ipv6/addrlabel.c index b055bc79f56d..c6311d7108f6 100644 --- a/net/ipv6/addrlabel.c +++ b/net/ipv6/addrlabel.c @@ -30,7 +30,6 @@ * Policy Table */ struct ip6addrlbl_entry { - possible_net_t lbl_net; struct in6_addr prefix; int prefixlen; int ifindex; @@ -41,19 +40,6 @@ struct ip6addrlbl_entry { struct rcu_head rcu; }; -static struct ip6addrlbl_table -{ - struct hlist_head head; - spinlock_t lock; - u32 seq; -} ip6addrlbl_table; - -static inline -struct net *ip6addrlbl_net(const struct ip6addrlbl_entry *lbl) -{ - return read_pnet(&lbl->lbl_net); -} - /* * Default policy table (RFC6724 + extensions) * @@ -148,13 +134,10 @@ static inline void ip6addrlbl_put(struct ip6addrlbl_entry *p) } /* Find label */ -static bool __ip6addrlbl_match(struct net *net, - const struct ip6addrlbl_entry *p, +static bool __ip6addrlbl_match(const struct ip6addrlbl_entry *p, const struct in6_addr *addr, int addrtype, int ifindex) { - if (!net_eq(ip6addrlbl_net(p), net)) - return false; if (p->ifindex && p->ifindex != ifindex) return false; if (p->addrtype && p->addrtype != addrtype) @@ -169,8 +152,9 @@ static struct ip6addrlbl_entry *__ipv6_addr_label(struct net *net, int type, int ifindex) { struct ip6addrlbl_entry *p; - hlist_for_each_entry_rcu(p, &ip6addrlbl_table.head, list) { - if (__ip6addrlbl_match(net, p, addr, type, ifindex)) + + hlist_for_each_entry_rcu(p, &net->ipv6.ip6addrlbl_table.head, list) { + if (__ip6addrlbl_match(p, addr, type, ifindex)) return p; } return NULL; @@ -196,8 +180,7 @@ u32 ipv6_addr_label(struct net *net, } /* allocate one entry */ -static struct ip6addrlbl_entry *ip6addrlbl_alloc(struct net *net, - const struct in6_addr *prefix, +static struct ip6addrlbl_entry *ip6addrlbl_alloc(const struct in6_addr *prefix, int prefixlen, int ifindex, u32 label) { @@ -236,24 +219,23 @@ static struct ip6addrlbl_entry *ip6addrlbl_alloc(struct net *net, newp->addrtype = addrtype; newp->label = label; INIT_HLIST_NODE(&newp->list); - write_pnet(&newp->lbl_net, net); refcount_set(&newp->refcnt, 1); return newp; } /* add a label */ -static int __ip6addrlbl_add(struct ip6addrlbl_entry *newp, int replace) +static int __ip6addrlbl_add(struct net *net, struct ip6addrlbl_entry *newp, + int replace) { - struct hlist_node *n; struct ip6addrlbl_entry *last = NULL, *p = NULL; + struct hlist_node *n; int ret = 0; ADDRLABEL(KERN_DEBUG "%s(newp=%p, replace=%d)\n", __func__, newp, replace); - hlist_for_each_entry_safe(p, n, &ip6addrlbl_table.head, list) { + hlist_for_each_entry_safe(p, n, &net->ipv6.ip6addrlbl_table.head, list) { if (p->prefixlen == newp->prefixlen && - net_eq(ip6addrlbl_net(p), ip6addrlbl_net(newp)) && p->ifindex == newp->ifindex && ipv6_addr_equal(&p->prefix, &newp->prefix)) { if (!replace) { @@ -273,10 +255,10 @@ static int __ip6addrlbl_add(struct ip6addrlbl_entry *newp, int replace) if (last) hlist_add_behind_rcu(&newp->list, &last->list); else - hlist_add_head_rcu(&newp->list, &ip6addrlbl_table.head); + hlist_add_head_rcu(&newp->list, &net->ipv6.ip6addrlbl_table.head); out: if (!ret) - ip6addrlbl_table.seq++; + net->ipv6.ip6addrlbl_table.seq++; return ret; } @@ -292,12 +274,12 @@ static int ip6addrlbl_add(struct net *net, __func__, prefix, prefixlen, ifindex, (unsigned int)label, replace); - newp = ip6addrlbl_alloc(net, prefix, prefixlen, ifindex, label); + newp = ip6addrlbl_alloc(prefix, prefixlen, ifindex, label); if (IS_ERR(newp)) return PTR_ERR(newp); - spin_lock(&ip6addrlbl_table.lock); - ret = __ip6addrlbl_add(newp, replace); - spin_unlock(&ip6addrlbl_table.lock); + spin_lock(&net->ipv6.ip6addrlbl_table.lock); + ret = __ip6addrlbl_add(net, newp, replace); + spin_unlock(&net->ipv6.ip6addrlbl_table.lock); if (ret) ip6addrlbl_free(newp); return ret; @@ -315,9 +297,8 @@ static int __ip6addrlbl_del(struct net *net, ADDRLABEL(KERN_DEBUG "%s(prefix=%pI6, prefixlen=%d, ifindex=%d)\n", __func__, prefix, prefixlen, ifindex); - hlist_for_each_entry_safe(p, n, &ip6addrlbl_table.head, list) { + hlist_for_each_entry_safe(p, n, &net->ipv6.ip6addrlbl_table.head, list) { if (p->prefixlen == prefixlen && - net_eq(ip6addrlbl_net(p), net) && p->ifindex == ifindex && ipv6_addr_equal(&p->prefix, prefix)) { hlist_del_rcu(&p->list); @@ -340,9 +321,9 @@ static int ip6addrlbl_del(struct net *net, __func__, prefix, prefixlen, ifindex); ipv6_addr_prefix(&prefix_buf, prefix, prefixlen); - spin_lock(&ip6addrlbl_table.lock); + spin_lock(&net->ipv6.ip6addrlbl_table.lock); ret = __ip6addrlbl_del(net, &prefix_buf, prefixlen, ifindex); - spin_unlock(&ip6addrlbl_table.lock); + spin_unlock(&net->ipv6.ip6addrlbl_table.lock); return ret; } @@ -354,6 +335,9 @@ static int __net_init ip6addrlbl_net_init(struct net *net) ADDRLABEL(KERN_DEBUG "%s\n", __func__); + spin_lock_init(&net->ipv6.ip6addrlbl_table.lock); + INIT_HLIST_HEAD(&net->ipv6.ip6addrlbl_table.head); + for (i = 0; i < ARRAY_SIZE(ip6addrlbl_init_table); i++) { int ret = ip6addrlbl_add(net, ip6addrlbl_init_table[i].prefix, @@ -373,14 +357,12 @@ static void __net_exit ip6addrlbl_net_exit(struct net *net) struct hlist_node *n; /* Remove all labels belonging to the exiting net */ - spin_lock(&ip6addrlbl_table.lock); - hlist_for_each_entry_safe(p, n, &ip6addrlbl_table.head, list) { - if (net_eq(ip6addrlbl_net(p), net)) { - hlist_del_rcu(&p->list); - ip6addrlbl_put(p); - } + spin_lock(&net->ipv6.ip6addrlbl_table.lock); + hlist_for_each_entry_safe(p, n, &net->ipv6.ip6addrlbl_table.head, list) { + hlist_del_rcu(&p->list); + ip6addrlbl_put(p); } - spin_unlock(&ip6addrlbl_table.lock); + spin_unlock(&net->ipv6.ip6addrlbl_table.lock); } static struct pernet_operations ipv6_addr_label_ops = { @@ -390,8 +372,6 @@ static struct pernet_operations ipv6_addr_label_ops = { int __init ipv6_addr_label_init(void) { - spin_lock_init(&ip6addrlbl_table.lock); - return register_pernet_subsys(&ipv6_addr_label_ops); } @@ -510,11 +490,10 @@ static int ip6addrlbl_dump(struct sk_buff *skb, struct netlink_callback *cb) int err; rcu_read_lock(); - hlist_for_each_entry_rcu(p, &ip6addrlbl_table.head, list) { - if (idx >= s_idx && - net_eq(ip6addrlbl_net(p), net)) { + hlist_for_each_entry_rcu(p, &net->ipv6.ip6addrlbl_table.head, list) { + if (idx >= s_idx) { err = ip6addrlbl_fill(skb, p, - ip6addrlbl_table.seq, + net->ipv6.ip6addrlbl_table.seq, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, RTM_NEWADDRLABEL, @@ -571,7 +550,7 @@ static int ip6addrlbl_get(struct sk_buff *in_skb, struct nlmsghdr *nlh, p = __ipv6_addr_label(net, addr, ipv6_addr_type(addr), ifal->ifal_index); if (p && !ip6addrlbl_hold(p)) p = NULL; - lseq = ip6addrlbl_table.seq; + lseq = net->ipv6.ip6addrlbl_table.seq; rcu_read_unlock(); if (!p) { -- cgit v1.2.3 From 64bc17811b72758753e2b64cd8f2a63812c61fe1 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 19 Sep 2017 16:27:09 -0700 Subject: ipv4: speedup ipv6 tunnels dismantle Implement exit_batch() method to dismantle more devices per round. (rtnl_lock() ... unregister_netdevice_many() ... rtnl_unlock()) Tested: $ cat add_del_unshare.sh for i in `seq 1 40` do (for j in `seq 1 100` ; do unshare -n /bin/true >/dev/null ; done) & done wait ; grep net_namespace /proc/slabinfo Before patch : $ time ./add_del_unshare.sh net_namespace 126 282 5504 1 2 : tunables 8 4 0 : slabdata 126 282 0 real 1m38.965s user 0m0.688s sys 0m37.017s After patch: $ time ./add_del_unshare.sh net_namespace 135 291 5504 1 2 : tunables 8 4 0 : slabdata 135 291 0 real 0m22.117s user 0m0.728s sys 0m35.328s Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/ip_tunnels.h | 3 ++- net/ipv4/ip_gre.c | 22 +++++++++------------- net/ipv4/ip_tunnel.c | 12 +++++++++--- net/ipv4/ip_vti.c | 7 +++---- net/ipv4/ipip.c | 7 +++---- 5 files changed, 26 insertions(+), 25 deletions(-) (limited to 'include/net') diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h index 992652856fe8..b41a1e057fce 100644 --- a/include/net/ip_tunnels.h +++ b/include/net/ip_tunnels.h @@ -258,7 +258,8 @@ int ip_tunnel_get_iflink(const struct net_device *dev); int ip_tunnel_init_net(struct net *net, unsigned int ip_tnl_net_id, struct rtnl_link_ops *ops, char *devname); -void ip_tunnel_delete_net(struct ip_tunnel_net *itn, struct rtnl_link_ops *ops); +void ip_tunnel_delete_nets(struct list_head *list_net, unsigned int id, + struct rtnl_link_ops *ops); void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev, const struct iphdr *tnl_params, const u8 protocol); diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index 0162fb955b33..9cee986ac6b8 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -1013,15 +1013,14 @@ static int __net_init ipgre_init_net(struct net *net) return ip_tunnel_init_net(net, ipgre_net_id, &ipgre_link_ops, NULL); } -static void __net_exit ipgre_exit_net(struct net *net) +static void __net_exit ipgre_exit_batch_net(struct list_head *list_net) { - struct ip_tunnel_net *itn = net_generic(net, ipgre_net_id); - ip_tunnel_delete_net(itn, &ipgre_link_ops); + ip_tunnel_delete_nets(list_net, ipgre_net_id, &ipgre_link_ops); } static struct pernet_operations ipgre_net_ops = { .init = ipgre_init_net, - .exit = ipgre_exit_net, + .exit_batch = ipgre_exit_batch_net, .id = &ipgre_net_id, .size = sizeof(struct ip_tunnel_net), }; @@ -1540,15 +1539,14 @@ static int __net_init ipgre_tap_init_net(struct net *net) return ip_tunnel_init_net(net, gre_tap_net_id, &ipgre_tap_ops, "gretap0"); } -static void __net_exit ipgre_tap_exit_net(struct net *net) +static void __net_exit ipgre_tap_exit_batch_net(struct list_head *list_net) { - struct ip_tunnel_net *itn = net_generic(net, gre_tap_net_id); - ip_tunnel_delete_net(itn, &ipgre_tap_ops); + ip_tunnel_delete_nets(list_net, gre_tap_net_id, &ipgre_tap_ops); } static struct pernet_operations ipgre_tap_net_ops = { .init = ipgre_tap_init_net, - .exit = ipgre_tap_exit_net, + .exit_batch = ipgre_tap_exit_batch_net, .id = &gre_tap_net_id, .size = sizeof(struct ip_tunnel_net), }; @@ -1559,16 +1557,14 @@ static int __net_init erspan_init_net(struct net *net) &erspan_link_ops, "erspan0"); } -static void __net_exit erspan_exit_net(struct net *net) +static void __net_exit erspan_exit_batch_net(struct list_head *net_list) { - struct ip_tunnel_net *itn = net_generic(net, erspan_net_id); - - ip_tunnel_delete_net(itn, &erspan_link_ops); + ip_tunnel_delete_nets(net_list, erspan_net_id, &erspan_link_ops); } static struct pernet_operations erspan_net_ops = { .init = erspan_init_net, - .exit = erspan_exit_net, + .exit_batch = erspan_exit_batch_net, .id = &erspan_net_id, .size = sizeof(struct ip_tunnel_net), }; diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index e9805ad664ac..fe6fee728ce4 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -1061,16 +1061,22 @@ static void ip_tunnel_destroy(struct ip_tunnel_net *itn, struct list_head *head, } } -void ip_tunnel_delete_net(struct ip_tunnel_net *itn, struct rtnl_link_ops *ops) +void ip_tunnel_delete_nets(struct list_head *net_list, unsigned int id, + struct rtnl_link_ops *ops) { + struct ip_tunnel_net *itn; + struct net *net; LIST_HEAD(list); rtnl_lock(); - ip_tunnel_destroy(itn, &list, ops); + list_for_each_entry(net, net_list, exit_list) { + itn = net_generic(net, id); + ip_tunnel_destroy(itn, &list, ops); + } unregister_netdevice_many(&list); rtnl_unlock(); } -EXPORT_SYMBOL_GPL(ip_tunnel_delete_net); +EXPORT_SYMBOL_GPL(ip_tunnel_delete_nets); int ip_tunnel_newlink(struct net_device *dev, struct nlattr *tb[], struct ip_tunnel_parm *p, __u32 fwmark) diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c index 5ed63d250950..02d70ca99db1 100644 --- a/net/ipv4/ip_vti.c +++ b/net/ipv4/ip_vti.c @@ -452,15 +452,14 @@ static int __net_init vti_init_net(struct net *net) return 0; } -static void __net_exit vti_exit_net(struct net *net) +static void __net_exit vti_exit_batch_net(struct list_head *list_net) { - struct ip_tunnel_net *itn = net_generic(net, vti_net_id); - ip_tunnel_delete_net(itn, &vti_link_ops); + ip_tunnel_delete_nets(list_net, vti_net_id, &vti_link_ops); } static struct pernet_operations vti_net_ops = { .init = vti_init_net, - .exit = vti_exit_net, + .exit_batch = vti_exit_batch_net, .id = &vti_net_id, .size = sizeof(struct ip_tunnel_net), }; diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index fb1ad22b5e29..1e47818e38c7 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -634,15 +634,14 @@ static int __net_init ipip_init_net(struct net *net) return ip_tunnel_init_net(net, ipip_net_id, &ipip_link_ops, "tunl0"); } -static void __net_exit ipip_exit_net(struct net *net) +static void __net_exit ipip_exit_batch_net(struct list_head *list_net) { - struct ip_tunnel_net *itn = net_generic(net, ipip_net_id); - ip_tunnel_delete_net(itn, &ipip_link_ops); + ip_tunnel_delete_nets(list_net, ipip_net_id, &ipip_link_ops); } static struct pernet_operations ipip_net_ops = { .init = ipip_init_net, - .exit = ipip_exit_net, + .exit_batch = ipip_exit_batch_net, .id = &ipip_net_id, .size = sizeof(struct ip_tunnel_net), }; -- cgit v1.2.3 From 2512b1b18d0748d867bb22387db7c86b903291ad Mon Sep 17 00:00:00 2001 From: Liad Kaufman Date: Sat, 5 Aug 2017 11:44:31 +0300 Subject: mac80211: extend ieee80211_ie_split to support EXTENSION Current ieee80211_ie_split() implementation doesn't account for elements that are sub-elements of the EXTENSION IE. To extend support to these IEs as well, treat the WLAN_EID_EXTENSION ids in the %ids array as indicating that the next id in the array is a sub-element of the EXTENSION IE. Signed-off-by: Liad Kaufman Signed-off-by: Luca Coelho Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 6 ++++-- net/wireless/util.c | 54 ++++++++++++++++++++++++++++++++++++++++++-------- 2 files changed, 50 insertions(+), 10 deletions(-) (limited to 'include/net') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index f12fa5245a45..aa9d993e519a 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -5934,7 +5934,8 @@ int cfg80211_get_p2p_attr(const u8 *ies, unsigned int len, * @ies: the IE buffer * @ielen: the length of the IE buffer * @ids: an array with element IDs that are allowed before - * the split + * the split. A WLAN_EID_EXTENSION value means that the next + * EID in the list is a sub-element of the EXTENSION IE. * @n_ids: the size of the element ID array * @after_ric: array IE types that come after the RIC element * @n_after_ric: size of the @after_ric array @@ -5965,7 +5966,8 @@ size_t ieee80211_ie_split_ric(const u8 *ies, size_t ielen, * @ies: the IE buffer * @ielen: the length of the IE buffer * @ids: an array with element IDs that are allowed before - * the split + * the split. A WLAN_EID_EXTENSION value means that the next + * EID in the list is a sub-element of the EXTENSION IE. * @n_ids: the size of the element ID array * @offset: offset where to start splitting in the buffer * diff --git a/net/wireless/util.c b/net/wireless/util.c index bcb1284c3415..4aab793c2f00 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -1367,13 +1367,29 @@ int cfg80211_get_p2p_attr(const u8 *ies, unsigned int len, } EXPORT_SYMBOL(cfg80211_get_p2p_attr); -static bool ieee80211_id_in_list(const u8 *ids, int n_ids, u8 id) +static bool ieee80211_id_in_list(const u8 *ids, int n_ids, u8 id, bool id_ext) { int i; - for (i = 0; i < n_ids; i++) - if (ids[i] == id) + /* Make sure array values are legal */ + if (WARN_ON(ids[n_ids - 1] == WLAN_EID_EXTENSION)) + return false; + + i = 0; + while (i < n_ids) { + if (ids[i] == WLAN_EID_EXTENSION) { + if (id_ext && (ids[i + 1] == id)) + return true; + + i += 2; + continue; + } + + if (ids[i] == id && !id_ext) return true; + + i++; + } return false; } @@ -1403,14 +1419,36 @@ size_t ieee80211_ie_split_ric(const u8 *ies, size_t ielen, { size_t pos = offset; - while (pos < ielen && ieee80211_id_in_list(ids, n_ids, ies[pos])) { + while (pos < ielen) { + u8 ext = 0; + + if (ies[pos] == WLAN_EID_EXTENSION) + ext = 2; + if ((pos + ext) >= ielen) + break; + + if (!ieee80211_id_in_list(ids, n_ids, ies[pos + ext], + ies[pos] == WLAN_EID_EXTENSION)) + break; + if (ies[pos] == WLAN_EID_RIC_DATA && n_after_ric) { pos = skip_ie(ies, ielen, pos); - while (pos < ielen && - !ieee80211_id_in_list(after_ric, n_after_ric, - ies[pos])) - pos = skip_ie(ies, ielen, pos); + while (pos < ielen) { + if (ies[pos] == WLAN_EID_EXTENSION) + ext = 2; + else + ext = 0; + + if ((pos + ext) >= ielen) + break; + + if (!ieee80211_id_in_list(after_ric, + n_after_ric, + ies[pos + ext], + ext == 2)) + pos = skip_ie(ies, ielen, pos); + } } else { pos = skip_ie(ies, ielen, pos); } -- cgit v1.2.3 From 1272c5d89b597995cb10db87dd4a1adc91d36006 Mon Sep 17 00:00:00 2001 From: Luca Coelho Date: Fri, 18 Aug 2017 15:33:56 +0300 Subject: mac80211: add documentation to ieee80211_rx_ba_offl() Add documentation to ieee80211_rx_ba_offl() function and, while at it, rename the bit argument to tid, for consistency. Signed-off-by: Luca Coelho Signed-off-by: Johannes Berg --- include/net/mac80211.h | 8 +++++++- net/mac80211/agg-rx.c | 4 ++-- 2 files changed, 9 insertions(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 885690fa39c8..cc9073e45be9 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -5441,8 +5441,14 @@ void ieee80211_mark_rx_ba_filtered_frames(struct ieee80211_sta *pubsta, u8 tid, */ void ieee80211_send_bar(struct ieee80211_vif *vif, u8 *ra, u16 tid, u16 ssn); +/** + * ieee80211_manage_rx_ba_offl - helper to queue an RX BA work + * @vif: &struct ieee80211_vif pointer from the add_interface callback + * @addr: station mac address + * @tid: the rx tid + */ void ieee80211_manage_rx_ba_offl(struct ieee80211_vif *vif, const u8 *addr, - unsigned int bit); + unsigned int tid); /** * ieee80211_start_rx_ba_session_offl - start a Rx BA session diff --git a/net/mac80211/agg-rx.c b/net/mac80211/agg-rx.c index 2849a1fc41c5..88cc1ae935ea 100644 --- a/net/mac80211/agg-rx.c +++ b/net/mac80211/agg-rx.c @@ -459,7 +459,7 @@ void ieee80211_process_addba_request(struct ieee80211_local *local, } void ieee80211_manage_rx_ba_offl(struct ieee80211_vif *vif, - const u8 *addr, unsigned int bit) + const u8 *addr, unsigned int tid) { struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif); struct ieee80211_local *local = sdata->local; @@ -470,7 +470,7 @@ void ieee80211_manage_rx_ba_offl(struct ieee80211_vif *vif, if (!sta) goto unlock; - set_bit(bit, sta->ampdu_mlme.tid_rx_manage_offl); + set_bit(tid, sta->ampdu_mlme.tid_rx_manage_offl); ieee80211_queue_work(&local->hw, &sta->ampdu_mlme.work); unlock: rcu_read_unlock(); -- cgit v1.2.3 From a6bcda44843c6dfced0fb973e2607c2a98addfa9 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 19 Sep 2017 11:52:43 +0200 Subject: cfg80211: remove unused function ieee80211_data_from_8023() This function hasn't been used since the removal of iwmc3200wifi in 2012. It also appears to have a bug when qos=True, since then it'll copy uninitialized stack memory to the SKB. Just remove the function entirely. Reported-by: Jouni Malinen Signed-off-by: Johannes Berg --- Documentation/driver-api/80211/cfg80211.rst | 3 - include/net/cfg80211.h | 13 ---- net/wireless/util.c | 115 ---------------------------- 3 files changed, 131 deletions(-) (limited to 'include/net') diff --git a/Documentation/driver-api/80211/cfg80211.rst b/Documentation/driver-api/80211/cfg80211.rst index 8ffac57e1f5b..eeab91b59457 100644 --- a/Documentation/driver-api/80211/cfg80211.rst +++ b/Documentation/driver-api/80211/cfg80211.rst @@ -299,9 +299,6 @@ Data path helpers .. kernel-doc:: include/net/cfg80211.h :functions: ieee80211_data_to_8023 -.. kernel-doc:: include/net/cfg80211.h - :functions: ieee80211_data_from_8023 - .. kernel-doc:: include/net/cfg80211.h :functions: ieee80211_amsdu_to_8023s diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index aa9d993e519a..cc1996081463 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -4346,19 +4346,6 @@ static inline int ieee80211_data_to_8023(struct sk_buff *skb, const u8 *addr, return ieee80211_data_to_8023_exthdr(skb, NULL, addr, iftype); } -/** - * ieee80211_data_from_8023 - convert an 802.3 frame to 802.11 - * @skb: the 802.3 frame - * @addr: the device MAC address - * @iftype: the virtual interface type - * @bssid: the network bssid (used only for iftype STATION and ADHOC) - * @qos: build 802.11 QoS data frame - * Return: 0 on success, or a negative error code. - */ -int ieee80211_data_from_8023(struct sk_buff *skb, const u8 *addr, - enum nl80211_iftype iftype, const u8 *bssid, - bool qos); - /** * ieee80211_amsdu_to_8023s - decode an IEEE 802.11n A-MSDU frame * diff --git a/net/wireless/util.c b/net/wireless/util.c index 4aab793c2f00..7dcdf67cba29 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -529,121 +529,6 @@ int ieee80211_data_to_8023_exthdr(struct sk_buff *skb, struct ethhdr *ehdr, } EXPORT_SYMBOL(ieee80211_data_to_8023_exthdr); -int ieee80211_data_from_8023(struct sk_buff *skb, const u8 *addr, - enum nl80211_iftype iftype, - const u8 *bssid, bool qos) -{ - struct ieee80211_hdr hdr; - u16 hdrlen, ethertype; - __le16 fc; - const u8 *encaps_data; - int encaps_len, skip_header_bytes; - int nh_pos, h_pos; - int head_need; - - if (unlikely(skb->len < ETH_HLEN)) - return -EINVAL; - - nh_pos = skb_network_header(skb) - skb->data; - h_pos = skb_transport_header(skb) - skb->data; - - /* convert Ethernet header to proper 802.11 header (based on - * operation mode) */ - ethertype = (skb->data[12] << 8) | skb->data[13]; - fc = cpu_to_le16(IEEE80211_FTYPE_DATA | IEEE80211_STYPE_DATA); - - switch (iftype) { - case NL80211_IFTYPE_AP: - case NL80211_IFTYPE_AP_VLAN: - case NL80211_IFTYPE_P2P_GO: - fc |= cpu_to_le16(IEEE80211_FCTL_FROMDS); - /* DA BSSID SA */ - memcpy(hdr.addr1, skb->data, ETH_ALEN); - memcpy(hdr.addr2, addr, ETH_ALEN); - memcpy(hdr.addr3, skb->data + ETH_ALEN, ETH_ALEN); - hdrlen = 24; - break; - case NL80211_IFTYPE_STATION: - case NL80211_IFTYPE_P2P_CLIENT: - fc |= cpu_to_le16(IEEE80211_FCTL_TODS); - /* BSSID SA DA */ - memcpy(hdr.addr1, bssid, ETH_ALEN); - memcpy(hdr.addr2, skb->data + ETH_ALEN, ETH_ALEN); - memcpy(hdr.addr3, skb->data, ETH_ALEN); - hdrlen = 24; - break; - case NL80211_IFTYPE_OCB: - case NL80211_IFTYPE_ADHOC: - /* DA SA BSSID */ - memcpy(hdr.addr1, skb->data, ETH_ALEN); - memcpy(hdr.addr2, skb->data + ETH_ALEN, ETH_ALEN); - memcpy(hdr.addr3, bssid, ETH_ALEN); - hdrlen = 24; - break; - default: - return -EOPNOTSUPP; - } - - if (qos) { - fc |= cpu_to_le16(IEEE80211_STYPE_QOS_DATA); - hdrlen += 2; - } - - hdr.frame_control = fc; - hdr.duration_id = 0; - hdr.seq_ctrl = 0; - - skip_header_bytes = ETH_HLEN; - if (ethertype == ETH_P_AARP || ethertype == ETH_P_IPX) { - encaps_data = bridge_tunnel_header; - encaps_len = sizeof(bridge_tunnel_header); - skip_header_bytes -= 2; - } else if (ethertype >= ETH_P_802_3_MIN) { - encaps_data = rfc1042_header; - encaps_len = sizeof(rfc1042_header); - skip_header_bytes -= 2; - } else { - encaps_data = NULL; - encaps_len = 0; - } - - skb_pull(skb, skip_header_bytes); - nh_pos -= skip_header_bytes; - h_pos -= skip_header_bytes; - - head_need = hdrlen + encaps_len - skb_headroom(skb); - - if (head_need > 0 || skb_cloned(skb)) { - head_need = max(head_need, 0); - if (head_need) - skb_orphan(skb); - - if (pskb_expand_head(skb, head_need, 0, GFP_ATOMIC)) - return -ENOMEM; - } - - if (encaps_data) { - memcpy(skb_push(skb, encaps_len), encaps_data, encaps_len); - nh_pos += encaps_len; - h_pos += encaps_len; - } - - memcpy(skb_push(skb, hdrlen), &hdr, hdrlen); - - nh_pos += hdrlen; - h_pos += hdrlen; - - /* Update skb pointers to various headers since this modified frame - * is going to go through Linux networking code that may potentially - * need things like pointer to IP header. */ - skb_reset_mac_header(skb); - skb_set_network_header(skb, nh_pos); - skb_set_transport_header(skb, h_pos); - - return 0; -} -EXPORT_SYMBOL(ieee80211_data_from_8023); - static void __frame_add_frag(struct sk_buff *skb, struct page *page, void *ptr, int len, int size) -- cgit v1.2.3 From 6e617de84e87d626d1e976fc30e1322239fd4d2d Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Wed, 20 Sep 2017 18:26:53 +0200 Subject: net: avoid a full fib lookup when rp_filter is disabled. Since commit 1dced6a85482 ("ipv4: Restore accept_local behaviour in fib_validate_source()") a full fib lookup is needed even if the rp_filter is disabled, if accept_local is false - which is the default. What we really need in the above scenario is just checking that the source IP address is not local, and in most case we can do that is a cheaper way looking up the ifaddr hash table. This commit adds a helper for such lookup, and uses it to validate the src address when rp_filter is disabled and no 'local' routes are created by the user space in the relevant namespace. A new ipv4 netns flag is added to account for such routes. We need that to preserve the same behavior we had before this patch. It also drops the checks to bail early from __fib_validate_source, added by the commit 1dced6a85482 ("ipv4: Restore accept_local behaviour in fib_validate_source()") they do not give any measurable performance improvement: if we do the lookup with are on a slower path. This improves UDP performances for unconnected sockets when rp_filter is disabled by 5% and also gives small but measurable performance improvement for TCP flood scenarios. v1 -> v2: - use the ifaddr lookup helper in __ip_dev_find(), as suggested by Eric - fall-back to full lookup if custom local routes are present Signed-off-by: Paolo Abeni Signed-off-by: David S. Miller --- include/linux/inetdevice.h | 1 + include/net/netns/ipv4.h | 1 + net/ipv4/devinet.c | 30 ++++++++++++++++++------------ net/ipv4/fib_frontend.c | 22 +++++++++++++++++----- 4 files changed, 37 insertions(+), 17 deletions(-) (limited to 'include/net') diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h index fb3f809e34e4..751d051f0bc7 100644 --- a/include/linux/inetdevice.h +++ b/include/linux/inetdevice.h @@ -179,6 +179,7 @@ __be32 inet_confirm_addr(struct net *net, struct in_device *in_dev, __be32 dst, __be32 local, int scope); struct in_ifaddr *inet_ifa_byprefix(struct in_device *in_dev, __be32 prefix, __be32 mask); +struct in_ifaddr *inet_lookup_ifaddr_rcu(struct net *net, __be32 addr); static __inline__ bool inet_ifa_match(__be32 addr, struct in_ifaddr *ifa) { return !((addr^ifa->ifa_address)&ifa->ifa_mask); diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 20d061c805e3..20720721da4b 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -49,6 +49,7 @@ struct netns_ipv4 { #ifdef CONFIG_IP_MULTIPLE_TABLES struct fib_rules_ops *rules_ops; bool fib_has_custom_rules; + bool fib_has_custom_local_routes; struct fib_table __rcu *fib_main; struct fib_table __rcu *fib_default; #endif diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index d7adc0616599..7ce22a2c07ce 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -137,22 +137,12 @@ static void inet_hash_remove(struct in_ifaddr *ifa) */ struct net_device *__ip_dev_find(struct net *net, __be32 addr, bool devref) { - u32 hash = inet_addr_hash(net, addr); struct net_device *result = NULL; struct in_ifaddr *ifa; rcu_read_lock(); - hlist_for_each_entry_rcu(ifa, &inet_addr_lst[hash], hash) { - if (ifa->ifa_local == addr) { - struct net_device *dev = ifa->ifa_dev->dev; - - if (!net_eq(dev_net(dev), net)) - continue; - result = dev; - break; - } - } - if (!result) { + ifa = inet_lookup_ifaddr_rcu(net, addr); + if (!ifa) { struct flowi4 fl4 = { .daddr = addr }; struct fib_result res = { 0 }; struct fib_table *local; @@ -165,6 +155,8 @@ struct net_device *__ip_dev_find(struct net *net, __be32 addr, bool devref) !fib_table_lookup(local, &fl4, &res, FIB_LOOKUP_NOREF) && res.type == RTN_LOCAL) result = FIB_RES_DEV(res); + } else { + result = ifa->ifa_dev->dev; } if (result && devref) dev_hold(result); @@ -173,6 +165,20 @@ struct net_device *__ip_dev_find(struct net *net, __be32 addr, bool devref) } EXPORT_SYMBOL(__ip_dev_find); +/* called under RCU lock */ +struct in_ifaddr *inet_lookup_ifaddr_rcu(struct net *net, __be32 addr) +{ + u32 hash = inet_addr_hash(net, addr); + struct in_ifaddr *ifa; + + hlist_for_each_entry_rcu(ifa, &inet_addr_lst[hash], hash) + if (ifa->ifa_local == addr && + net_eq(dev_net(ifa->ifa_dev->dev), net)) + return ifa; + + return NULL; +} + static void rtmsg_ifa(int event, struct in_ifaddr *, struct nlmsghdr *, u32); static BLOCKING_NOTIFIER_HEAD(inetaddr_chain); diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 37819ab4cc74..f02819134ba2 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -345,9 +345,6 @@ static int __fib_validate_source(struct sk_buff *skb, __be32 src, __be32 dst, if (res.type != RTN_UNICAST && (res.type != RTN_LOCAL || !IN_DEV_ACCEPT_LOCAL(idev))) goto e_inval; - if (!rpf && !fib_num_tclassid_users(net) && - (dev->ifindex != oif || !IN_DEV_TX_REDIRECTS(idev))) - goto last_resort; fib_combine_itag(itag, &res); dev_match = false; @@ -402,13 +399,26 @@ int fib_validate_source(struct sk_buff *skb, __be32 src, __be32 dst, struct in_device *idev, u32 *itag) { int r = secpath_exists(skb) ? 0 : IN_DEV_RPFILTER(idev); + struct net *net = dev_net(dev); - if (!r && !fib_num_tclassid_users(dev_net(dev)) && - IN_DEV_ACCEPT_LOCAL(idev) && + if (!r && !fib_num_tclassid_users(net) && (dev->ifindex != oif || !IN_DEV_TX_REDIRECTS(idev))) { + if (IN_DEV_ACCEPT_LOCAL(idev)) + goto ok; + /* if no local routes are added from user space we can check + * for local addresses looking-up the ifaddr table + */ + if (net->ipv4.fib_has_custom_local_routes) + goto full_check; + if (inet_lookup_ifaddr_rcu(net, src)) + return -EINVAL; + +ok: *itag = 0; return 0; } + +full_check: return __fib_validate_source(skb, src, dst, tos, oif, dev, r, idev, itag); } @@ -759,6 +769,8 @@ static int inet_rtm_newroute(struct sk_buff *skb, struct nlmsghdr *nlh, } err = fib_table_insert(net, tb, &cfg, extack); + if (!err && cfg.fc_type == RTN_LOCAL) + net->ipv4.fib_has_custom_local_routes = true; errout: return err; } -- cgit v1.2.3 From a1f3316dd7b5ce740c774697c664e2e60d095794 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 21 Sep 2017 18:18:23 -0700 Subject: ipv4: Move fib_has_custom_local_routes outside of IP_MULTIPLE_TABLES. > net/ipv4/fib_frontend.c: In function 'fib_validate_source': > net/ipv4/fib_frontend.c:411:16: error: 'struct netns_ipv4' has no member named 'fib_has_custom_local_routes' > if (net->ipv4.fib_has_custom_local_routes) > ^ > net/ipv4/fib_frontend.c: In function 'inet_rtm_newroute': > net/ipv4/fib_frontend.c:773:12: error: 'struct netns_ipv4' has no member named 'fib_has_custom_local_routes' > net->ipv4.fib_has_custom_local_routes = true; > ^ Fixes: 6e617de84e87 ("net: avoid a full fib lookup when rp_filter is disabled.") Reported-by: Stephen Rothwell Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 20720721da4b..8387f099115e 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -49,10 +49,10 @@ struct netns_ipv4 { #ifdef CONFIG_IP_MULTIPLE_TABLES struct fib_rules_ops *rules_ops; bool fib_has_custom_rules; - bool fib_has_custom_local_routes; struct fib_table __rcu *fib_main; struct fib_table __rcu *fib_default; #endif + bool fib_has_custom_local_routes; #ifdef CONFIG_IP_ROUTE_CLASSID int fib_num_tclassid_users; #endif -- cgit v1.2.3 From 373b8eeb0c15d4ce58f62afb12f213b1b5bbc3d3 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 21 Sep 2017 23:45:43 +0300 Subject: xfrm: make aead_len() return unsigned int Key lengths can't be negative. Comparison with nla_len() is left signed just in case negative value can sneak in there. Signed-off-by: Alexey Dobriyan Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 2 +- net/xfrm/xfrm_user.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index f002a2c5e33c..0be4c547e383 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1764,7 +1764,7 @@ static inline int xfrm_acquire_is_on(struct net *net) } #endif -static inline int aead_len(struct xfrm_algo_aead *alg) +static inline unsigned int aead_len(struct xfrm_algo_aead *alg) { return sizeof(*alg) + ((alg->alg_key_len + 7) / 8); } diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 2bfbd9121e3b..32c67b80c3ce 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -84,7 +84,7 @@ static int verify_aead(struct nlattr **attrs) return 0; algp = nla_data(rt); - if (nla_len(rt) < aead_len(algp)) + if (nla_len(rt) < (int)aead_len(algp)) return -EINVAL; algp->alg_name[sizeof(algp->alg_name) - 1] = '\0'; -- cgit v1.2.3 From 06cd22f830f28023b82455c82c7db65fc6cf9c16 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 21 Sep 2017 23:46:30 +0300 Subject: xfrm: make xfrm_alg_len() return unsigned int Key lengths can't be negative. Comparison with nla_len() is left signed just in case negative value can sneak in there. Signed-off-by: Alexey Dobriyan Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 2 +- net/xfrm/xfrm_user.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 0be4c547e383..2abc0e117f11 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1769,7 +1769,7 @@ static inline unsigned int aead_len(struct xfrm_algo_aead *alg) return sizeof(*alg) + ((alg->alg_key_len + 7) / 8); } -static inline int xfrm_alg_len(const struct xfrm_algo *alg) +static inline unsigned int xfrm_alg_len(const struct xfrm_algo *alg) { return sizeof(*alg) + ((alg->alg_key_len + 7) / 8); } diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 32c67b80c3ce..09512d90e6a5 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -42,7 +42,7 @@ static int verify_one_alg(struct nlattr **attrs, enum xfrm_attr_type_t type) return 0; algp = nla_data(rt); - if (nla_len(rt) < xfrm_alg_len(algp)) + if (nla_len(rt) < (int)xfrm_alg_len(algp)) return -EINVAL; switch (type) { -- cgit v1.2.3 From 1bd963a72e859d194d87a5a2a8839efee7e23102 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 21 Sep 2017 23:47:09 +0300 Subject: xfrm: make xfrm_alg_auth_len() return unsigned int Key lengths can't be negative. Comparison with nla_len() is left signed just in case negative value can sneak in there. Signed-off-by: Alexey Dobriyan Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 2 +- net/xfrm/xfrm_user.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 2abc0e117f11..5d5e11b653eb 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1774,7 +1774,7 @@ static inline unsigned int xfrm_alg_len(const struct xfrm_algo *alg) return sizeof(*alg) + ((alg->alg_key_len + 7) / 8); } -static inline int xfrm_alg_auth_len(const struct xfrm_algo_auth *alg) +static inline unsigned int xfrm_alg_auth_len(const struct xfrm_algo_auth *alg) { return sizeof(*alg) + ((alg->alg_key_len + 7) / 8); } diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 09512d90e6a5..465c23d4ea78 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -68,7 +68,7 @@ static int verify_auth_trunc(struct nlattr **attrs) return 0; algp = nla_data(rt); - if (nla_len(rt) < xfrm_alg_auth_len(algp)) + if (nla_len(rt) < (int)xfrm_alg_auth_len(algp)) return -EINVAL; algp->alg_name[sizeof(algp->alg_name) - 1] = '\0'; -- cgit v1.2.3 From 5e708e47c44366453c33373940455a75fd33f635 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 21 Sep 2017 23:47:50 +0300 Subject: xfrm: make xfrm_replay_state_esn_len() return unsigned int Replay detection bitmaps can't have negative length. Comparisons with nla_len() are left signed just in case negative value can sneak in there. Propagate unsignedness for code size savings: add/remove: 0/0 grow/shrink: 0/5 up/down: 0/-38 (-38) function old new delta xfrm_state_construct 1802 1800 -2 xfrm_update_ae_params 295 289 -6 xfrm_state_migrate 1345 1339 -6 xfrm_replay_notify_esn 349 337 -12 xfrm_replay_notify_bmp 345 333 -12 Signed-off-by: Alexey Dobriyan Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 2 +- net/xfrm/xfrm_user.c | 10 +++++----- 2 files changed, 6 insertions(+), 6 deletions(-) (limited to 'include/net') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 5d5e11b653eb..3cb618bbcfa5 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1779,7 +1779,7 @@ static inline unsigned int xfrm_alg_auth_len(const struct xfrm_algo_auth *alg) return sizeof(*alg) + ((alg->alg_key_len + 7) / 8); } -static inline int xfrm_replay_state_esn_len(struct xfrm_replay_state_esn *replay_esn) +static inline unsigned int xfrm_replay_state_esn_len(struct xfrm_replay_state_esn *replay_esn) { return sizeof(*replay_esn) + replay_esn->bmp_len * sizeof(__u32); } diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 465c23d4ea78..83718db5ec9c 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -130,7 +130,7 @@ static inline int verify_replay(struct xfrm_usersa_info *p, if (rs->bmp_len > XFRMA_REPLAY_ESN_MAX / sizeof(rs->bmp[0]) / 8) return -EINVAL; - if (nla_len(rt) < xfrm_replay_state_esn_len(rs) && + if (nla_len(rt) < (int)xfrm_replay_state_esn_len(rs) && nla_len(rt) != sizeof(*rs)) return -EINVAL; } @@ -404,7 +404,7 @@ static inline int xfrm_replay_verify_len(struct xfrm_replay_state_esn *replay_es struct nlattr *rp) { struct xfrm_replay_state_esn *up; - int ulen; + unsigned int ulen; if (!replay_esn || !rp) return 0; @@ -414,7 +414,7 @@ static inline int xfrm_replay_verify_len(struct xfrm_replay_state_esn *replay_es /* Check the overall length and the internal bitmap length to avoid * potential overflow. */ - if (nla_len(rp) < ulen || + if (nla_len(rp) < (int)ulen || xfrm_replay_state_esn_len(replay_esn) != ulen || replay_esn->bmp_len != up->bmp_len) return -EINVAL; @@ -430,14 +430,14 @@ static int xfrm_alloc_replay_state_esn(struct xfrm_replay_state_esn **replay_esn struct nlattr *rta) { struct xfrm_replay_state_esn *p, *pp, *up; - int klen, ulen; + unsigned int klen, ulen; if (!rta) return 0; up = nla_data(rta); klen = xfrm_replay_state_esn_len(up); - ulen = nla_len(rta) >= klen ? klen : sizeof(*up); + ulen = nla_len(rta) >= (int)klen ? klen : sizeof(*up); p = kzalloc(klen, GFP_KERNEL); if (!p) -- cgit v1.2.3 From e451ae8e4f6b3f6bd3b83a5595657b5421b3bf69 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Sat, 23 Sep 2017 23:01:06 +0300 Subject: neigh: make struct neigh_table::entry_size unsigned int Neigh entry size can't be negative. Space savings: add/remove: 0/0 grow/shrink: 0/5 up/down: 0/-7 (-7) function old new delta lowpan_neigh_construct 25 24 -1 clip_seq_sub_iter 152 151 -1 clip_ioctl 1475 1474 -1 clip_constructor 93 92 -1 __neigh_create 2455 2452 -3 Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- include/net/neighbour.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/neighbour.h b/include/net/neighbour.h index 9816df225af3..9a25512e0a6e 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -190,7 +190,7 @@ struct neigh_hash_table { struct neigh_table { int family; - int entry_size; + unsigned int entry_size; int key_len; __be16 protocol; __u32 (*hash)(const void *pkey, -- cgit v1.2.3 From 01ccdf126ca5f9d4fe0889f65ee67afac910f19c Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Sat, 23 Sep 2017 23:03:04 +0300 Subject: neigh: make strucrt neigh_table::entry_size unsigned int Key length can't be negative. Leave comparisons against nla_len() signed just in case truncated attribute can sneak in there. Space savings: add/remove: 0/0 grow/shrink: 0/7 up/down: 0/-7 (-7) function old new delta pneigh_delete 273 272 -1 mlx5e_rep_netevent_event 1415 1414 -1 mlx5e_create_encap_header_ipv6 1194 1193 -1 mlx5e_create_encap_header_ipv4 1071 1070 -1 cxgb4_l2t_get 1104 1103 -1 __pneigh_lookup 69 68 -1 __neigh_create 2452 2451 -1 Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- drivers/net/ethernet/chelsio/cxgb4/l2t.c | 4 ++-- include/net/neighbour.h | 2 +- net/core/neighbour.c | 18 +++++++++--------- 3 files changed, 12 insertions(+), 12 deletions(-) (limited to 'include/net') diff --git a/drivers/net/ethernet/chelsio/cxgb4/l2t.c b/drivers/net/ethernet/chelsio/cxgb4/l2t.c index f7ef8871dd0b..1817a0307d26 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/l2t.c +++ b/drivers/net/ethernet/chelsio/cxgb4/l2t.c @@ -422,7 +422,7 @@ struct l2t_entry *cxgb4_l2t_get(struct l2t_data *d, struct neighbour *neigh, u8 lport; u16 vlan; struct l2t_entry *e; - int addr_len = neigh->tbl->key_len; + unsigned int addr_len = neigh->tbl->key_len; u32 *addr = (u32 *)neigh->primary_key; int ifidx = neigh->dev->ifindex; int hash = addr_hash(d, addr, addr_len, ifidx); @@ -536,7 +536,7 @@ void t4_l2t_update(struct adapter *adap, struct neighbour *neigh) struct l2t_entry *e; struct sk_buff_head *arpq = NULL; struct l2t_data *d = adap->l2t; - int addr_len = neigh->tbl->key_len; + unsigned int addr_len = neigh->tbl->key_len; u32 *addr = (u32 *) neigh->primary_key; int ifidx = neigh->dev->ifindex; int hash = addr_hash(d, addr, addr_len, ifidx); diff --git a/include/net/neighbour.h b/include/net/neighbour.h index 9a25512e0a6e..2492000e1035 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -191,7 +191,7 @@ struct neigh_hash_table { struct neigh_table { int family; unsigned int entry_size; - int key_len; + unsigned int key_len; __be16 protocol; __u32 (*hash)(const void *pkey, const struct net_device *dev, diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 16a1a4c4eb57..6ea3a1a7f36a 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -457,7 +457,7 @@ struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, struct net *net, const void *pkey) { struct neighbour *n; - int key_len = tbl->key_len; + unsigned int key_len = tbl->key_len; u32 hash_val; struct neigh_hash_table *nht; @@ -488,7 +488,7 @@ struct neighbour *__neigh_create(struct neigh_table *tbl, const void *pkey, struct net_device *dev, bool want_ref) { u32 hash_val; - int key_len = tbl->key_len; + unsigned int key_len = tbl->key_len; int error; struct neighbour *n1, *rc, *n = neigh_alloc(tbl, dev); struct neigh_hash_table *nht; @@ -572,7 +572,7 @@ out_neigh_release: } EXPORT_SYMBOL(__neigh_create); -static u32 pneigh_hash(const void *pkey, int key_len) +static u32 pneigh_hash(const void *pkey, unsigned int key_len) { u32 hash_val = *(u32 *)(pkey + key_len - 4); hash_val ^= (hash_val >> 16); @@ -585,7 +585,7 @@ static u32 pneigh_hash(const void *pkey, int key_len) static struct pneigh_entry *__pneigh_lookup_1(struct pneigh_entry *n, struct net *net, const void *pkey, - int key_len, + unsigned int key_len, struct net_device *dev) { while (n) { @@ -601,7 +601,7 @@ static struct pneigh_entry *__pneigh_lookup_1(struct pneigh_entry *n, struct pneigh_entry *__pneigh_lookup(struct neigh_table *tbl, struct net *net, const void *pkey, struct net_device *dev) { - int key_len = tbl->key_len; + unsigned int key_len = tbl->key_len; u32 hash_val = pneigh_hash(pkey, key_len); return __pneigh_lookup_1(tbl->phash_buckets[hash_val], @@ -614,7 +614,7 @@ struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl, struct net_device *dev, int creat) { struct pneigh_entry *n; - int key_len = tbl->key_len; + unsigned int key_len = tbl->key_len; u32 hash_val = pneigh_hash(pkey, key_len); read_lock_bh(&tbl->lock); @@ -659,7 +659,7 @@ int pneigh_delete(struct neigh_table *tbl, struct net *net, const void *pkey, struct net_device *dev) { struct pneigh_entry *n, **np; - int key_len = tbl->key_len; + unsigned int key_len = tbl->key_len; u32 hash_val = pneigh_hash(pkey, key_len); write_lock_bh(&tbl->lock); @@ -1662,7 +1662,7 @@ static int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, if (tbl == NULL) return -EAFNOSUPPORT; - if (nla_len(dst_attr) < tbl->key_len) + if (nla_len(dst_attr) < (int)tbl->key_len) goto out; if (ndm->ndm_flags & NTF_PROXY) { @@ -1730,7 +1730,7 @@ static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, if (tbl == NULL) return -EAFNOSUPPORT; - if (nla_len(tb[NDA_DST]) < tbl->key_len) + if (nla_len(tb[NDA_DST]) < (int)tbl->key_len) goto out; dst = nla_data(tb[NDA_DST]); lladdr = tb[NDA_LLADDR] ? nla_data(tb[NDA_LLADDR]) : NULL; -- cgit v1.2.3 From 3b8e9238a8d194e82f0202a5fb68a63686ebe420 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Mon, 25 Sep 2017 10:58:21 +0200 Subject: net: sched: introduce helper to identify gact pass action Introduce a helper called is_tcf_gact_pass which could be used to tell if the action is gact pass or not. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/tc_act/tc_gact.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/net') diff --git a/include/net/tc_act/tc_gact.h b/include/net/tc_act/tc_gact.h index 41afe1ce7b16..d979a0d48f9e 100644 --- a/include/net/tc_act/tc_gact.h +++ b/include/net/tc_act/tc_gact.h @@ -33,6 +33,11 @@ static inline bool __is_tcf_gact_act(const struct tc_action *a, int act, return false; } +static inline bool is_tcf_gact_ok(const struct tc_action *a) +{ + return __is_tcf_gact_act(a, TC_ACT_OK, false); +} + static inline bool is_tcf_gact_shot(const struct tc_action *a) { return __is_tcf_gact_act(a, TC_ACT_SHOT, false); -- cgit v1.2.3 From 85e482285bbbd508483cbe08de69c8fe00cdbbfe Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Wed, 27 Sep 2017 08:23:11 +0200 Subject: fib: notifier: Add VIF add and delete event types In order for an interface to forward packets according to the kernel multicast routing table, it must be configured with a VIF index according to the mroute user API. The VIF index is then used to refer to that interface in the mroute user API, for example, to set the iif and oifs of an MFC entry. In order to allow drivers to be aware and offload multicast routes, they have to be aware of the VIF add and delete notifications. Due to the fact that a specific VIF can be deleted and re-added pointing to another netdevice, and the MFC routes that point to it will forward the matching packets to the new netdevice, a driver willing to offload MFC cache entries must be aware of the VIF add and delete events in addition to MFC routes notifications. Signed-off-by: Yotam Gigi Reviewed-by: Ido Schimmel Signed-off-by: Jiri Pirko Reviewed-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/net/fib_notifier.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/net') diff --git a/include/net/fib_notifier.h b/include/net/fib_notifier.h index 669b9716dc7a..54cd6b839d2f 100644 --- a/include/net/fib_notifier.h +++ b/include/net/fib_notifier.h @@ -20,6 +20,8 @@ enum fib_event_type { FIB_EVENT_RULE_DEL, FIB_EVENT_NH_ADD, FIB_EVENT_NH_DEL, + FIB_EVENT_VIF_ADD, + FIB_EVENT_VIF_DEL, }; struct fib_notifier_ops { -- cgit v1.2.3 From 4d65b9487831170e699b2fc64a91b839d729bd78 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Wed, 27 Sep 2017 08:23:13 +0200 Subject: ipmr: Add FIB notification access functions Make the ipmr module register as a FIB notifier. To do that, implement both the ipmr_seq_read and ipmr_dump ops. The ipmr_seq_read op returns a sequence counter that is incremented on every notification related operation done by the ipmr. To implement that, add a sequence counter in the netns_ipv4 struct and increment it whenever a new MFC route or VIF are added or deleted. The sequence operations are protected by the RTNL lock. The ipmr_dump iterates the list of MFC routes and the list of VIF entries and sends notifications about them. The entries dump is done under RCU where the VIF dump uses the mrt_lock too, as the vif->dev field can change under RCU. Signed-off-by: Yotam Gigi Reviewed-by: Ido Schimmel Signed-off-by: Jiri Pirko Reviewed-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/linux/mroute.h | 15 ++++++ include/net/netns/ipv4.h | 3 ++ net/ipv4/ipmr.c | 137 ++++++++++++++++++++++++++++++++++++++++++++++- 3 files changed, 153 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/linux/mroute.h b/include/linux/mroute.h index 10028f208efb..54c5cb82ddcb 100644 --- a/include/linux/mroute.h +++ b/include/linux/mroute.h @@ -5,6 +5,7 @@ #include #include #include +#include #include #ifdef CONFIG_IP_MROUTE @@ -58,6 +59,14 @@ struct vif_device { int link; /* Physical interface index */ }; +struct vif_entry_notifier_info { + struct fib_notifier_info info; + struct net_device *dev; + vifi_t vif_index; + unsigned short vif_flags; + u32 tb_id; +}; + #define VIFF_STATIC 0x8000 #define VIF_EXISTS(_mrt, _idx) ((_mrt)->vif_table[_idx].dev != NULL) @@ -146,6 +155,12 @@ struct mfc_cache { struct rcu_head rcu; }; +struct mfc_entry_notifier_info { + struct fib_notifier_info info; + struct mfc_cache *mfc; + u32 tb_id; +}; + struct rtmsg; int ipmr_get_route(struct net *net, struct sk_buff *skb, __be32 saddr, __be32 daddr, diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 8387f099115e..abc84d986da4 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -163,6 +163,9 @@ struct netns_ipv4 { struct fib_notifier_ops *notifier_ops; unsigned int fib_seq; /* protected by rtnl_mutex */ + struct fib_notifier_ops *ipmr_notifier_ops; + unsigned int ipmr_seq; /* protected by rtnl_mutex */ + atomic_t rt_genid; }; #endif diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 86dc5f98c5dd..49879c338357 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -264,6 +264,16 @@ static void __net_exit ipmr_rules_exit(struct net *net) fib_rules_unregister(net->ipv4.mr_rules_ops); rtnl_unlock(); } + +static int ipmr_rules_dump(struct net *net, struct notifier_block *nb) +{ + return fib_rules_dump(net, nb, RTNL_FAMILY_IPMR); +} + +static unsigned int ipmr_rules_seq_read(struct net *net) +{ + return fib_rules_seq_read(net, RTNL_FAMILY_IPMR); +} #else #define ipmr_for_each_table(mrt, net) \ for (mrt = net->ipv4.mrt; mrt; mrt = NULL) @@ -298,6 +308,16 @@ static void __net_exit ipmr_rules_exit(struct net *net) net->ipv4.mrt = NULL; rtnl_unlock(); } + +static int ipmr_rules_dump(struct net *net, struct notifier_block *nb) +{ + return 0; +} + +static unsigned int ipmr_rules_seq_read(struct net *net) +{ + return 0; +} #endif static inline int ipmr_hash_cmp(struct rhashtable_compare_arg *arg, @@ -587,6 +607,43 @@ static struct net_device *ipmr_reg_vif(struct net *net, struct mr_table *mrt) } #endif +static int call_ipmr_vif_entry_notifier(struct notifier_block *nb, + struct net *net, + enum fib_event_type event_type, + struct vif_device *vif, + vifi_t vif_index, u32 tb_id) +{ + struct vif_entry_notifier_info info = { + .info = { + .family = RTNL_FAMILY_IPMR, + .net = net, + }, + .dev = vif->dev, + .vif_index = vif_index, + .vif_flags = vif->flags, + .tb_id = tb_id, + }; + + return call_fib_notifier(nb, net, event_type, &info.info); +} + +static int call_ipmr_mfc_entry_notifier(struct notifier_block *nb, + struct net *net, + enum fib_event_type event_type, + struct mfc_cache *mfc, u32 tb_id) +{ + struct mfc_entry_notifier_info info = { + .info = { + .family = RTNL_FAMILY_IPMR, + .net = net, + }, + .mfc = mfc, + .tb_id = tb_id + }; + + return call_fib_notifier(nb, net, event_type, &info.info); +} + /** * vif_delete - Delete a VIF entry * @notify: Set to 1, if the caller is a notifier_call @@ -3050,14 +3107,87 @@ static const struct net_protocol pim_protocol = { }; #endif +static unsigned int ipmr_seq_read(struct net *net) +{ + ASSERT_RTNL(); + + return net->ipv4.ipmr_seq + ipmr_rules_seq_read(net); +} + +static int ipmr_dump(struct net *net, struct notifier_block *nb) +{ + struct mr_table *mrt; + int err; + + err = ipmr_rules_dump(net, nb); + if (err) + return err; + + ipmr_for_each_table(mrt, net) { + struct vif_device *v = &mrt->vif_table[0]; + struct mfc_cache *mfc; + int vifi; + + /* Notifiy on table VIF entries */ + read_lock(&mrt_lock); + for (vifi = 0; vifi < mrt->maxvif; vifi++, v++) { + if (!v->dev) + continue; + + call_ipmr_vif_entry_notifier(nb, net, FIB_EVENT_VIF_ADD, + v, vifi, mrt->id); + } + read_unlock(&mrt_lock); + + /* Notify on table MFC entries */ + list_for_each_entry_rcu(mfc, &mrt->mfc_cache_list, list) + call_ipmr_mfc_entry_notifier(nb, net, + FIB_EVENT_ENTRY_ADD, mfc, + mrt->id); + } + + return 0; +} + +static const struct fib_notifier_ops ipmr_notifier_ops_template = { + .family = RTNL_FAMILY_IPMR, + .fib_seq_read = ipmr_seq_read, + .fib_dump = ipmr_dump, + .owner = THIS_MODULE, +}; + +int __net_init ipmr_notifier_init(struct net *net) +{ + struct fib_notifier_ops *ops; + + net->ipv4.ipmr_seq = 0; + + ops = fib_notifier_ops_register(&ipmr_notifier_ops_template, net); + if (IS_ERR(ops)) + return PTR_ERR(ops); + net->ipv4.ipmr_notifier_ops = ops; + + return 0; +} + +static void __net_exit ipmr_notifier_exit(struct net *net) +{ + fib_notifier_ops_unregister(net->ipv4.ipmr_notifier_ops); + net->ipv4.ipmr_notifier_ops = NULL; +} + /* Setup for IP multicast routing */ static int __net_init ipmr_net_init(struct net *net) { int err; + err = ipmr_notifier_init(net); + if (err) + goto ipmr_notifier_fail; + err = ipmr_rules_init(net); if (err < 0) - goto fail; + goto ipmr_rules_fail; #ifdef CONFIG_PROC_FS err = -ENOMEM; @@ -3074,7 +3204,9 @@ proc_cache_fail: proc_vif_fail: ipmr_rules_exit(net); #endif -fail: +ipmr_rules_fail: + ipmr_notifier_exit(net); +ipmr_notifier_fail: return err; } @@ -3084,6 +3216,7 @@ static void __net_exit ipmr_net_exit(struct net *net) remove_proc_entry("ip_mr_cache", net->proc_net); remove_proc_entry("ip_mr_vif", net->proc_net); #endif + ipmr_notifier_exit(net); ipmr_rules_exit(net); } -- cgit v1.2.3 From c7c3e5913bf18eda3cf38932bebdce48351baac9 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Wed, 27 Sep 2017 19:08:00 -0700 Subject: net: ipv4: remove fib_weight fib_weight in fib_info is set but not used. Remove it and the helpers for setting it. Signed-off-by: David Ahern Signed-off-by: David S. Miller --- include/net/ip_fib.h | 3 --- net/ipv4/fib_semantics.c | 9 --------- 2 files changed, 12 deletions(-) (limited to 'include/net') diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h index 1a7f7e424320..f80524396c06 100644 --- a/include/net/ip_fib.h +++ b/include/net/ip_fib.h @@ -122,9 +122,6 @@ struct fib_info { #define fib_rtt fib_metrics->metrics[RTAX_RTT-1] #define fib_advmss fib_metrics->metrics[RTAX_ADVMSS-1] int fib_nhs; -#ifdef CONFIG_IP_ROUTE_MULTIPATH - int fib_weight; -#endif struct rcu_head rcu; struct fib_nh fib_nh[0]; #define fib_dev fib_nh[0].nh_dev diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 57a5d48acee8..be0874620ecc 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -601,17 +601,9 @@ static void fib_rebalance(struct fib_info *fi) atomic_set(&nexthop_nh->nh_upper_bound, upper_bound); } endfor_nexthops(fi); } - -static inline void fib_add_weight(struct fib_info *fi, - const struct fib_nh *nh) -{ - fi->fib_weight += nh->nh_weight; -} - #else /* CONFIG_IP_ROUTE_MULTIPATH */ #define fib_rebalance(fi) do { } while (0) -#define fib_add_weight(fi, nh) do { } while (0) #endif /* CONFIG_IP_ROUTE_MULTIPATH */ @@ -1275,7 +1267,6 @@ struct fib_info *fib_create_info(struct fib_config *cfg, change_nexthops(fi) { fib_info_update_nh_saddr(net, nexthop_nh); - fib_add_weight(fi, nexthop_nh); } endfor_nexthops(fi) fib_rebalance(fi); -- cgit v1.2.3 From 152402483ed75b167d5628d414e876ffa7a6d4c4 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Fri, 29 Sep 2017 17:19:18 -0400 Subject: net: dsa: add tagging ops to port The DSA tagging protocol operations are specific to each CPU port, thus the dsa_device_ops pointer belongs to the dsa_port structure. >From now on assign a slave's xmit copy from its CPU port tagging operations. This will ease the future support for multiple CPU ports. Also keep the tag_ops at the beginning of the dsa_port structure so that we ensure copies for hot path are in cacheline 1. Signed-off-by: Vivien Didelot Signed-off-by: David S. Miller --- include/net/dsa.h | 3 +++ net/dsa/dsa2.c | 1 + net/dsa/dsa_priv.h | 2 +- net/dsa/legacy.c | 1 + net/dsa/slave.c | 3 +-- 5 files changed, 7 insertions(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 8dee216a5a9b..4d1df2f086e8 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -175,6 +175,9 @@ struct dsa_mall_tc_entry { struct dsa_port { + /* CPU port tagging operations used by master or slave devices */ + const struct dsa_device_ops *tag_ops; + struct dsa_switch *ds; unsigned int index; const char *name; diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index 6a10c5c1639f..9eac4726dc0c 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -522,6 +522,7 @@ static int dsa_cpu_parse(struct dsa_port *port, u32 index, return PTR_ERR(tag_ops); } + dst->cpu_dp->tag_ops = tag_ops; dst->tag_ops = tag_ops; dst->rcv = dst->tag_ops->rcv; diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h index d429505dc4e7..9397291bb3aa 100644 --- a/net/dsa/dsa_priv.h +++ b/net/dsa/dsa_priv.h @@ -66,7 +66,7 @@ struct dsa_notifier_vlan_info { }; struct dsa_slave_priv { - /* Copy of dp->ds->dst->tag_ops->xmit for faster access in hot path */ + /* Copy of CPU port xmit for faster access in slave transmit hot path */ struct sk_buff * (*xmit)(struct sk_buff *skb, struct net_device *dev); diff --git a/net/dsa/legacy.c b/net/dsa/legacy.c index 8e849013f69d..4d374541815a 100644 --- a/net/dsa/legacy.c +++ b/net/dsa/legacy.c @@ -152,6 +152,7 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, if (IS_ERR(tag_ops)) return PTR_ERR(tag_ops); + dst->cpu_dp->tag_ops = tag_ops; dst->tag_ops = tag_ops; dst->rcv = dst->tag_ops->rcv; } diff --git a/net/dsa/slave.c b/net/dsa/slave.c index bf8800de13c1..4b634db05cee 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -1117,7 +1117,6 @@ int dsa_slave_resume(struct net_device *slave_dev) int dsa_slave_create(struct dsa_port *port, const char *name) { struct dsa_switch *ds = port->ds; - struct dsa_switch_tree *dst = ds->dst; struct net_device *master; struct net_device *slave_dev; struct dsa_slave_priv *p; @@ -1162,7 +1161,7 @@ int dsa_slave_create(struct dsa_port *port, const char *name) } p->dp = port; INIT_LIST_HEAD(&p->mall_tc_list); - p->xmit = dst->tag_ops->xmit; + p->xmit = cpu_dp->tag_ops->xmit; p->old_pause = -1; p->old_link = -1; -- cgit v1.2.3 From 3e41f93b358a8800194b87995ad076fc50919719 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Fri, 29 Sep 2017 17:19:19 -0400 Subject: net: dsa: prepare master receive hot path In preparation to make DSA master devices point to their corresponding CPU port instead of the whole tree, add copies of dst and rcv in the dsa_port structure so that we keep fast access in the receive hot path. Also keep the copies at the beginning of the dsa_port structure in order to ensure they are available in cacheline 1. Signed-off-by: Vivien Didelot Signed-off-by: David S. Miller --- include/net/dsa.h | 5 +++++ net/dsa/dsa2.c | 4 ++++ net/dsa/legacy.c | 4 ++++ 3 files changed, 13 insertions(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 4d1df2f086e8..6bda01fa5747 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -178,6 +178,11 @@ struct dsa_port { /* CPU port tagging operations used by master or slave devices */ const struct dsa_device_ops *tag_ops; + /* Copies for faster access in master receive hot path */ + struct dsa_switch_tree *dst; + struct sk_buff *(*rcv)(struct sk_buff *skb, struct net_device *dev, + struct packet_type *pt); + struct dsa_switch *ds; unsigned int index; const char *name; diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index 9eac4726dc0c..b71e3bb478e4 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -524,7 +524,11 @@ static int dsa_cpu_parse(struct dsa_port *port, u32 index, dst->cpu_dp->tag_ops = tag_ops; dst->tag_ops = tag_ops; + + /* Make a few copies for faster access in master receive hot path */ + dst->cpu_dp->rcv = dst->cpu_dp->tag_ops->rcv; dst->rcv = dst->tag_ops->rcv; + dst->cpu_dp->dst = dst; return 0; } diff --git a/net/dsa/legacy.c b/net/dsa/legacy.c index 4d374541815a..96c7e3f8b8bb 100644 --- a/net/dsa/legacy.c +++ b/net/dsa/legacy.c @@ -154,7 +154,11 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, dst->cpu_dp->tag_ops = tag_ops; dst->tag_ops = tag_ops; + + /* Few copies for faster access in master receive hot path */ + dst->cpu_dp->rcv = dst->cpu_dp->tag_ops->rcv; dst->rcv = dst->tag_ops->rcv; + dst->cpu_dp->dst = dst; } memcpy(ds->rtable, cd->rtable, sizeof(ds->rtable)); -- cgit v1.2.3 From aa193d9b1d7ea6893ce24a9d141f676950563987 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Fri, 29 Sep 2017 17:19:21 -0400 Subject: net: dsa: remove tag ops from the switch tree Now that the dsa_ptr is a dsa_port instance, there is no need to keep the tag operations in the dsa_switch_tree structure. Remove it. Signed-off-by: Vivien Didelot Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- include/net/dsa.h | 11 ----------- net/dsa/dsa2.c | 2 -- net/dsa/legacy.c | 2 -- 3 files changed, 15 deletions(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 6bda01fa5747..10dceccd9ce8 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -130,11 +130,6 @@ struct dsa_switch_tree { */ struct dsa_platform_data *pd; - /* Copy of tag_ops->rcv for faster access in hot path */ - struct sk_buff * (*rcv)(struct sk_buff *skb, - struct net_device *dev, - struct packet_type *pt); - /* * The switch port to which the CPU is attached. */ @@ -144,12 +139,6 @@ struct dsa_switch_tree { * Data for the individual switch chips. */ struct dsa_switch *ds[DSA_MAX_SWITCHES]; - - /* - * Tagging protocol operations for adding and removing an - * encapsulation tag. - */ - const struct dsa_device_ops *tag_ops; }; /* TC matchall action types, only mirroring for now */ diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index 62302558f38c..54ed054777bd 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -523,11 +523,9 @@ static int dsa_cpu_parse(struct dsa_port *port, u32 index, } dst->cpu_dp->tag_ops = tag_ops; - dst->tag_ops = tag_ops; /* Make a few copies for faster access in master receive hot path */ dst->cpu_dp->rcv = dst->cpu_dp->tag_ops->rcv; - dst->rcv = dst->tag_ops->rcv; dst->cpu_dp->dst = dst; return 0; diff --git a/net/dsa/legacy.c b/net/dsa/legacy.c index 71917505a5cc..19ff6e0a21dc 100644 --- a/net/dsa/legacy.c +++ b/net/dsa/legacy.c @@ -153,11 +153,9 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, return PTR_ERR(tag_ops); dst->cpu_dp->tag_ops = tag_ops; - dst->tag_ops = tag_ops; /* Few copies for faster access in master receive hot path */ dst->cpu_dp->rcv = dst->cpu_dp->tag_ops->rcv; - dst->rcv = dst->tag_ops->rcv; dst->cpu_dp->dst = dst; } -- cgit v1.2.3 From e1cfcbe82b4534bd0f99fef92a6d33843fd85e0e Mon Sep 17 00:00:00 2001 From: Haishuang Yan Date: Wed, 27 Sep 2017 11:35:40 +0800 Subject: ipv4: Namespaceify tcp_fastopen knob Different namespace application might require enable TCP Fast Open feature independently of the host. This patch series continues making more of the TCP Fast Open related sysctl knobs be per net-namespace. Reported-by: Luca BRUNO Signed-off-by: Haishuang Yan Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/af_inet.c | 7 ++++--- net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp.c | 4 ++-- net/ipv4/tcp_fastopen.c | 11 +++++------ net/ipv4/tcp_ipv4.c | 2 ++ 7 files changed, 21 insertions(+), 19 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index abc84d986da4..16420ccaef15 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -128,6 +128,7 @@ struct netns_ipv4 { int sysctl_tcp_timestamps; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; + int sysctl_tcp_fastopen; #ifdef CONFIG_NET_L3_MASTER_DEV int sysctl_udp_l3mdev_accept; diff --git a/include/net/tcp.h b/include/net/tcp.h index 770b608c8439..9e414a99034f 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -240,7 +240,6 @@ void tcp_time_wait(struct sock *sk, int state, int timeo); /* sysctl variables for tcp */ -extern int sysctl_tcp_fastopen; extern int sysctl_tcp_retrans_collapse; extern int sysctl_tcp_stdurg; extern int sysctl_tcp_rfc1337; diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index e31108e5ef79..ddd126d120ac 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -195,7 +195,7 @@ int inet_listen(struct socket *sock, int backlog) { struct sock *sk = sock->sk; unsigned char old_state; - int err; + int err, tcp_fastopen; lock_sock(sk); @@ -217,8 +217,9 @@ int inet_listen(struct socket *sock, int backlog) * because the socket was in TCP_LISTEN state previously but * was shutdown() rather than close(). */ - if ((sysctl_tcp_fastopen & TFO_SERVER_WO_SOCKOPT1) && - (sysctl_tcp_fastopen & TFO_SERVER_ENABLE) && + tcp_fastopen = sock_net(sk)->ipv4.sysctl_tcp_fastopen; + if ((tcp_fastopen & TFO_SERVER_WO_SOCKOPT1) && + (tcp_fastopen & TFO_SERVER_ENABLE) && !inet_csk(sk)->icsk_accept_queue.fastopenq.max_qlen) { fastopen_queue_tune(sk, backlog); tcp_fastopen_init_key_once(true); diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 0d3c038d7b04..e31e853cf486 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -400,13 +400,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "tcp_fastopen", - .data = &sysctl_tcp_fastopen, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec, - }, { .procname = "tcp_fastopen_key", .mode = 0600, @@ -1085,6 +1078,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_fastopen", + .data = &init_net.ipv4.sysctl_tcp_fastopen, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, #ifdef CONFIG_IP_ROUTE_MULTIPATH { .procname = "fib_multipath_use_neigh", diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 5091402720ab..dac56c4ad357 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1126,7 +1126,7 @@ static int tcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg, struct sockaddr *uaddr = msg->msg_name; int err, flags; - if (!(sysctl_tcp_fastopen & TFO_CLIENT_ENABLE) || + if (!(sock_net(sk)->ipv4.sysctl_tcp_fastopen & TFO_CLIENT_ENABLE) || (uaddr && msg->msg_namelen >= sizeof(uaddr->sa_family) && uaddr->sa_family == AF_UNSPEC)) return -EOPNOTSUPP; @@ -2759,7 +2759,7 @@ static int do_tcp_setsockopt(struct sock *sk, int level, case TCP_FASTOPEN_CONNECT: if (val > 1 || val < 0) { err = -EINVAL; - } else if (sysctl_tcp_fastopen & TFO_CLIENT_ENABLE) { + } else if (net->ipv4.sysctl_tcp_fastopen & TFO_CLIENT_ENABLE) { if (sk->sk_state == TCP_CLOSE) tp->fastopen_connect = val; else diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index e3c33220c418..31b08ec38cb8 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -9,8 +9,6 @@ #include #include -int sysctl_tcp_fastopen __read_mostly = TFO_CLIENT_ENABLE; - struct tcp_fastopen_context __rcu *tcp_fastopen_ctx; static DEFINE_SPINLOCK(tcp_fastopen_ctx_lock); @@ -279,21 +277,22 @@ struct sock *tcp_try_fastopen(struct sock *sk, struct sk_buff *skb, struct request_sock *req, struct tcp_fastopen_cookie *foc) { - struct tcp_fastopen_cookie valid_foc = { .len = -1 }; bool syn_data = TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq + 1; + int tcp_fastopen = sock_net(sk)->ipv4.sysctl_tcp_fastopen; + struct tcp_fastopen_cookie valid_foc = { .len = -1 }; struct sock *child; if (foc->len == 0) /* Client requests a cookie */ NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPFASTOPENCOOKIEREQD); - if (!((sysctl_tcp_fastopen & TFO_SERVER_ENABLE) && + if (!((tcp_fastopen & TFO_SERVER_ENABLE) && (syn_data || foc->len >= 0) && tcp_fastopen_queue_check(sk))) { foc->len = -1; return NULL; } - if (syn_data && (sysctl_tcp_fastopen & TFO_SERVER_COOKIE_NOT_REQD)) + if (syn_data && (tcp_fastopen & TFO_SERVER_COOKIE_NOT_REQD)) goto fastopen; if (foc->len >= 0 && /* Client presents or requests a cookie */ @@ -347,7 +346,7 @@ bool tcp_fastopen_cookie_check(struct sock *sk, u16 *mss, return false; } - if (sysctl_tcp_fastopen & TFO_CLIENT_NO_COOKIE) { + if (sock_net(sk)->ipv4.sysctl_tcp_fastopen & TFO_CLIENT_NO_COOKIE) { cookie->len = -1; return true; } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index d9416b5162bc..88409b13c9d2 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2472,6 +2472,8 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_window_scaling = 1; net->ipv4.sysctl_tcp_timestamps = 1; + net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; + return 0; fail: tcp_sk_exit(net); -- cgit v1.2.3 From dd000598a39b6937fcefdf143720ec9fb5250e72 Mon Sep 17 00:00:00 2001 From: Haishuang Yan Date: Wed, 27 Sep 2017 11:35:41 +0800 Subject: ipv4: Remove the 'publish' logic in tcp_fastopen_init_key_once MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The 'publish' logic is not necessary after commit dfea2aa65424 ("tcp: Do not call tcp_fastopen_reset_cipher from interrupt context"), because in tcp_fastopen_cookie_gen，it wouldn't call tcp_fastopen_init_key_once. Signed-off-by: Haishuang Yan Signed-off-by: David S. Miller --- include/net/tcp.h | 2 +- net/ipv4/af_inet.c | 2 +- net/ipv4/sysctl_net_ipv4.c | 5 ----- net/ipv4/tcp.c | 2 +- net/ipv4/tcp_fastopen.c | 4 ++-- 5 files changed, 5 insertions(+), 10 deletions(-) (limited to 'include/net') diff --git a/include/net/tcp.h b/include/net/tcp.h index 9e414a99034f..d9376e2458e9 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1555,7 +1555,7 @@ void tcp_fastopen_add_skb(struct sock *sk, struct sk_buff *skb); struct sock *tcp_try_fastopen(struct sock *sk, struct sk_buff *skb, struct request_sock *req, struct tcp_fastopen_cookie *foc); -void tcp_fastopen_init_key_once(bool publish); +void tcp_fastopen_init_key_once(void); bool tcp_fastopen_cookie_check(struct sock *sk, u16 *mss, struct tcp_fastopen_cookie *cookie); bool tcp_fastopen_defer_connect(struct sock *sk, int *err); diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index ddd126d120ac..e73ce79d7176 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -222,7 +222,7 @@ int inet_listen(struct socket *sock, int backlog) (tcp_fastopen & TFO_SERVER_ENABLE) && !inet_csk(sk)->icsk_accept_queue.fastopenq.max_qlen) { fastopen_queue_tune(sk, backlog); - tcp_fastopen_init_key_once(true); + tcp_fastopen_init_key_once(); } err = inet_csk_listen_start(sk, backlog); diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index e31e853cf486..f6324ead0e19 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -282,11 +282,6 @@ static int proc_tcp_fastopen_key(struct ctl_table *ctl, int write, ret = -EINVAL; goto bad_key; } - /* Generate a dummy secret but don't publish it. This - * is needed so we don't regenerate a new key on the - * first invocation of tcp_fastopen_cookie_gen - */ - tcp_fastopen_init_key_once(false); tcp_fastopen_reset_cipher(user_key, TCP_FASTOPEN_KEY_LENGTH); } diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index dac56c4ad357..4e395452d69f 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2749,7 +2749,7 @@ static int do_tcp_setsockopt(struct sock *sk, int level, case TCP_FASTOPEN: if (val >= 0 && ((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN))) { - tcp_fastopen_init_key_once(true); + tcp_fastopen_init_key_once(); fastopen_queue_tune(sk, val); } else { diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 31b08ec38cb8..8c8f0f0af59d 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -13,7 +13,7 @@ struct tcp_fastopen_context __rcu *tcp_fastopen_ctx; static DEFINE_SPINLOCK(tcp_fastopen_ctx_lock); -void tcp_fastopen_init_key_once(bool publish) +void tcp_fastopen_init_key_once(void) { static u8 key[TCP_FASTOPEN_KEY_LENGTH]; @@ -23,7 +23,7 @@ void tcp_fastopen_init_key_once(bool publish) * All call sites of tcp_fastopen_cookie_gen also check * for a valid cookie, so this is an acceptable risk. */ - if (net_get_random_once(key, sizeof(key)) && publish) + if (net_get_random_once(key, sizeof(key))) tcp_fastopen_reset_cipher(key, sizeof(key)); } -- cgit v1.2.3 From 437138485656c41e32b8c63c0987cfa0348be0e6 Mon Sep 17 00:00:00 2001 From: Haishuang Yan Date: Wed, 27 Sep 2017 11:35:42 +0800 Subject: ipv4: Namespaceify tcp_fastopen_key knob Different namespace application might require different tcp_fastopen_key independently of the host. David Miller pointed out there is a leak without releasing the context of tcp_fastopen_key during netns teardown. So add the release action in exit_batch path. Tested: 1. Container namespace: # cat /proc/sys/net/ipv4/tcp_fastopen_key: 2817fff2-f803cf97-eadfd1f3-78c0992b cookie key in tcp syn packets: Fast Open Cookie Kind: TCP Fast Open Cookie (34) Length: 10 Fast Open Cookie: 1e5dd82a8c492ca9 2. Host: # cat /proc/sys/net/ipv4/tcp_fastopen_key: 107d7c5f-68eb2ac7-02fb06e6-ed341702 cookie key in tcp syn packets: Fast Open Cookie Kind: TCP Fast Open Cookie (34) Length: 10 Fast Open Cookie: e213c02bf0afbc8a Signed-off-by: Haishuang Yan Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 4 +++ include/net/tcp.h | 6 ++--- net/ipv4/af_inet.c | 2 +- net/ipv4/sysctl_net_ipv4.c | 21 ++++++++------- net/ipv4/tcp.c | 2 +- net/ipv4/tcp_fastopen.c | 64 +++++++++++++++++++++++++++++++--------------- net/ipv4/tcp_ipv4.c | 6 +++++ 7 files changed, 70 insertions(+), 35 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 16420ccaef15..7bb9603ff66c 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -36,6 +36,8 @@ struct inet_timewait_death_row { int sysctl_max_tw_buckets; }; +struct tcp_fastopen_context; + struct netns_ipv4 { #ifdef CONFIG_SYSCTL struct ctl_table_header *forw_hdr; @@ -129,6 +131,8 @@ struct netns_ipv4 { struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; + struct tcp_fastopen_context __rcu *tcp_fastopen_ctx; + spinlock_t tcp_fastopen_ctx_lock; #ifdef CONFIG_NET_L3_MASTER_DEV int sysctl_udp_l3mdev_accept; diff --git a/include/net/tcp.h b/include/net/tcp.h index d9376e2458e9..6d25d8305054 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1549,13 +1549,13 @@ struct tcp_fastopen_request { }; void tcp_free_fastopen_req(struct tcp_sock *tp); -extern struct tcp_fastopen_context __rcu *tcp_fastopen_ctx; -int tcp_fastopen_reset_cipher(void *key, unsigned int len); +void tcp_fastopen_ctx_destroy(struct net *net); +int tcp_fastopen_reset_cipher(struct net *net, void *key, unsigned int len); void tcp_fastopen_add_skb(struct sock *sk, struct sk_buff *skb); struct sock *tcp_try_fastopen(struct sock *sk, struct sk_buff *skb, struct request_sock *req, struct tcp_fastopen_cookie *foc); -void tcp_fastopen_init_key_once(void); +void tcp_fastopen_init_key_once(struct net *net); bool tcp_fastopen_cookie_check(struct sock *sk, u16 *mss, struct tcp_fastopen_cookie *cookie); bool tcp_fastopen_defer_connect(struct sock *sk, int *err); diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index e73ce79d7176..43a1bbed7a42 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -222,7 +222,7 @@ int inet_listen(struct socket *sock, int backlog) (tcp_fastopen & TFO_SERVER_ENABLE) && !inet_csk(sk)->icsk_accept_queue.fastopenq.max_qlen) { fastopen_queue_tune(sk, backlog); - tcp_fastopen_init_key_once(); + tcp_fastopen_init_key_once(sock_net(sk)); } err = inet_csk_listen_start(sk, backlog); diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index f6324ead0e19..20e19fe78dbd 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -251,10 +251,12 @@ static int proc_allowed_congestion_control(struct ctl_table *ctl, return ret; } -static int proc_tcp_fastopen_key(struct ctl_table *ctl, int write, +static int proc_tcp_fastopen_key(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { + struct net *net = container_of(table->data, struct net, + ipv4.sysctl_tcp_fastopen); struct ctl_table tbl = { .maxlen = (TCP_FASTOPEN_KEY_LENGTH * 2 + 10) }; struct tcp_fastopen_context *ctxt; int ret; @@ -265,7 +267,7 @@ static int proc_tcp_fastopen_key(struct ctl_table *ctl, int write, return -ENOMEM; rcu_read_lock(); - ctxt = rcu_dereference(tcp_fastopen_ctx); + ctxt = rcu_dereference(net->ipv4.tcp_fastopen_ctx); if (ctxt) memcpy(user_key, ctxt->key, TCP_FASTOPEN_KEY_LENGTH); else @@ -282,7 +284,7 @@ static int proc_tcp_fastopen_key(struct ctl_table *ctl, int write, ret = -EINVAL; goto bad_key; } - tcp_fastopen_reset_cipher(user_key, TCP_FASTOPEN_KEY_LENGTH); + tcp_fastopen_reset_cipher(net, user_key, TCP_FASTOPEN_KEY_LENGTH); } bad_key: @@ -395,12 +397,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "tcp_fastopen_key", - .mode = 0600, - .maxlen = ((TCP_FASTOPEN_KEY_LENGTH * 2) + 10), - .proc_handler = proc_tcp_fastopen_key, - }, { .procname = "tcp_fastopen_blackhole_timeout_sec", .data = &sysctl_tcp_fastopen_blackhole_timeout, @@ -1080,6 +1076,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, + { + .procname = "tcp_fastopen_key", + .mode = 0600, + .data = &init_net.ipv4.sysctl_tcp_fastopen, + .maxlen = ((TCP_FASTOPEN_KEY_LENGTH * 2) + 10), + .proc_handler = proc_tcp_fastopen_key, + }, #ifdef CONFIG_IP_ROUTE_MULTIPATH { .procname = "fib_multipath_use_neigh", diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 4e395452d69f..23225c98d287 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2749,7 +2749,7 @@ static int do_tcp_setsockopt(struct sock *sk, int level, case TCP_FASTOPEN: if (val >= 0 && ((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN))) { - tcp_fastopen_init_key_once(); + tcp_fastopen_init_key_once(net); fastopen_queue_tune(sk, val); } else { diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 8c8f0f0af59d..4eae44ac3cb0 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -9,13 +9,18 @@ #include #include -struct tcp_fastopen_context __rcu *tcp_fastopen_ctx; - -static DEFINE_SPINLOCK(tcp_fastopen_ctx_lock); - -void tcp_fastopen_init_key_once(void) +void tcp_fastopen_init_key_once(struct net *net) { - static u8 key[TCP_FASTOPEN_KEY_LENGTH]; + u8 key[TCP_FASTOPEN_KEY_LENGTH]; + struct tcp_fastopen_context *ctxt; + + rcu_read_lock(); + ctxt = rcu_dereference(net->ipv4.tcp_fastopen_ctx); + if (ctxt) { + rcu_read_unlock(); + return; + } + rcu_read_unlock(); /* tcp_fastopen_reset_cipher publishes the new context * atomically, so we allow this race happening here. @@ -23,8 +28,8 @@ void tcp_fastopen_init_key_once(void) * All call sites of tcp_fastopen_cookie_gen also check * for a valid cookie, so this is an acceptable risk. */ - if (net_get_random_once(key, sizeof(key))) - tcp_fastopen_reset_cipher(key, sizeof(key)); + get_random_bytes(key, sizeof(key)); + tcp_fastopen_reset_cipher(net, key, sizeof(key)); } static void tcp_fastopen_ctx_free(struct rcu_head *head) @@ -35,7 +40,22 @@ static void tcp_fastopen_ctx_free(struct rcu_head *head) kfree(ctx); } -int tcp_fastopen_reset_cipher(void *key, unsigned int len) +void tcp_fastopen_ctx_destroy(struct net *net) +{ + struct tcp_fastopen_context *ctxt; + + spin_lock(&net->ipv4.tcp_fastopen_ctx_lock); + + ctxt = rcu_dereference_protected(net->ipv4.tcp_fastopen_ctx, + lockdep_is_held(&net->ipv4.tcp_fastopen_ctx_lock)); + rcu_assign_pointer(net->ipv4.tcp_fastopen_ctx, NULL); + spin_unlock(&net->ipv4.tcp_fastopen_ctx_lock); + + if (ctxt) + call_rcu(&ctxt->rcu, tcp_fastopen_ctx_free); +} + +int tcp_fastopen_reset_cipher(struct net *net, void *key, unsigned int len) { int err; struct tcp_fastopen_context *ctx, *octx; @@ -59,26 +79,27 @@ error: kfree(ctx); } memcpy(ctx->key, key, len); - spin_lock(&tcp_fastopen_ctx_lock); + spin_lock(&net->ipv4.tcp_fastopen_ctx_lock); - octx = rcu_dereference_protected(tcp_fastopen_ctx, - lockdep_is_held(&tcp_fastopen_ctx_lock)); - rcu_assign_pointer(tcp_fastopen_ctx, ctx); - spin_unlock(&tcp_fastopen_ctx_lock); + octx = rcu_dereference_protected(net->ipv4.tcp_fastopen_ctx, + lockdep_is_held(&net->ipv4.tcp_fastopen_ctx_lock)); + rcu_assign_pointer(net->ipv4.tcp_fastopen_ctx, ctx); + spin_unlock(&net->ipv4.tcp_fastopen_ctx_lock); if (octx) call_rcu(&octx->rcu, tcp_fastopen_ctx_free); return err; } -static bool __tcp_fastopen_cookie_gen(const void *path, +static bool __tcp_fastopen_cookie_gen(struct net *net, + const void *path, struct tcp_fastopen_cookie *foc) { struct tcp_fastopen_context *ctx; bool ok = false; rcu_read_lock(); - ctx = rcu_dereference(tcp_fastopen_ctx); + ctx = rcu_dereference(net->ipv4.tcp_fastopen_ctx); if (ctx) { crypto_cipher_encrypt_one(ctx->tfm, foc->val, path); foc->len = TCP_FASTOPEN_COOKIE_SIZE; @@ -94,7 +115,8 @@ static bool __tcp_fastopen_cookie_gen(const void *path, * * XXX (TFO) - refactor when TCP_FASTOPEN_COOKIE_SIZE != AES_BLOCK_SIZE. */ -static bool tcp_fastopen_cookie_gen(struct request_sock *req, +static bool tcp_fastopen_cookie_gen(struct net *net, + struct request_sock *req, struct sk_buff *syn, struct tcp_fastopen_cookie *foc) { @@ -102,7 +124,7 @@ static bool tcp_fastopen_cookie_gen(struct request_sock *req, const struct iphdr *iph = ip_hdr(syn); __be32 path[4] = { iph->saddr, iph->daddr, 0, 0 }; - return __tcp_fastopen_cookie_gen(path, foc); + return __tcp_fastopen_cookie_gen(net, path, foc); } #if IS_ENABLED(CONFIG_IPV6) @@ -110,13 +132,13 @@ static bool tcp_fastopen_cookie_gen(struct request_sock *req, const struct ipv6hdr *ip6h = ipv6_hdr(syn); struct tcp_fastopen_cookie tmp; - if (__tcp_fastopen_cookie_gen(&ip6h->saddr, &tmp)) { + if (__tcp_fastopen_cookie_gen(net, &ip6h->saddr, &tmp)) { struct in6_addr *buf = &tmp.addr; int i; for (i = 0; i < 4; i++) buf->s6_addr32[i] ^= ip6h->daddr.s6_addr32[i]; - return __tcp_fastopen_cookie_gen(buf, foc); + return __tcp_fastopen_cookie_gen(net, buf, foc); } } #endif @@ -296,7 +318,7 @@ struct sock *tcp_try_fastopen(struct sock *sk, struct sk_buff *skb, goto fastopen; if (foc->len >= 0 && /* Client presents or requests a cookie */ - tcp_fastopen_cookie_gen(req, skb, &valid_foc) && + tcp_fastopen_cookie_gen(sock_net(sk), req, skb, &valid_foc) && foc->len == TCP_FASTOPEN_COOKIE_SIZE && foc->len == valid_foc.len && !memcmp(foc->val, valid_foc.val, foc->len)) { diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 88409b13c9d2..49c74c0d0d21 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2473,6 +2473,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_timestamps = 1; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; + spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); return 0; fail: @@ -2483,7 +2484,12 @@ fail: static void __net_exit tcp_sk_exit_batch(struct list_head *net_exit_list) { + struct net *net; + inet_twsk_purge(&tcp_hashinfo, AF_INET); + + list_for_each_entry(net, net_exit_list, exit_list) + tcp_fastopen_ctx_destroy(net); } static struct pernet_operations __net_initdata tcp_sk_ops = { -- cgit v1.2.3 From 3733be14a32bae288b61ed28341e593baba983af Mon Sep 17 00:00:00 2001 From: Haishuang Yan Date: Wed, 27 Sep 2017 11:35:43 +0800 Subject: ipv4: Namespaceify tcp_fastopen_blackhole_timeout knob Different namespace application might require different time period in second to disable Fastopen on active TCP sockets. Tested: Simulate following similar situation that the server's data gets dropped after 3WHS. C ---- syn-data ---> S C <--- syn/ack ----- S C ---- ack --------> S S (accept & write) C? X <- data ------ S [retry and timeout] And then print netstat of TCPFastOpenBlackhole, the counter increased as expected when the firewall blackhole issue is detected and active TFO is disabled. # cat /proc/net/netstat | awk '{print $91}' TCPFastOpenBlackhole 1 Signed-off-by: Haishuang Yan Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 3 +++ net/ipv4/sysctl_net_ipv4.c | 20 +++++++++++--------- net/ipv4/tcp_fastopen.c | 30 +++++++++++------------------- net/ipv4/tcp_ipv4.c | 2 ++ 4 files changed, 27 insertions(+), 28 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 7bb9603ff66c..2c4222a5d102 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -133,6 +133,9 @@ struct netns_ipv4 { int sysctl_tcp_fastopen; struct tcp_fastopen_context __rcu *tcp_fastopen_ctx; spinlock_t tcp_fastopen_ctx_lock; + unsigned int sysctl_tcp_fastopen_blackhole_timeout; + atomic_t tfo_active_disable_times; + unsigned long tfo_active_disable_stamp; #ifdef CONFIG_NET_L3_MASTER_DEV int sysctl_udp_l3mdev_accept; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 20e19fe78dbd..cac8dd309f39 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -355,11 +355,13 @@ static int proc_tfo_blackhole_detect_timeout(struct ctl_table *table, void __user *buffer, size_t *lenp, loff_t *ppos) { + struct net *net = container_of(table->data, struct net, + ipv4.sysctl_tcp_fastopen_blackhole_timeout); int ret; ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); if (write && ret == 0) - tcp_fastopen_active_timeout_reset(); + atomic_set(&net->ipv4.tfo_active_disable_times, 0); return ret; } @@ -397,14 +399,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "tcp_fastopen_blackhole_timeout_sec", - .data = &sysctl_tcp_fastopen_blackhole_timeout, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_tfo_blackhole_detect_timeout, - .extra1 = &zero, - }, { .procname = "tcp_abort_on_overflow", .data = &sysctl_tcp_abort_on_overflow, @@ -1083,6 +1077,14 @@ static struct ctl_table ipv4_net_table[] = { .maxlen = ((TCP_FASTOPEN_KEY_LENGTH * 2) + 10), .proc_handler = proc_tcp_fastopen_key, }, + { + .procname = "tcp_fastopen_blackhole_timeout_sec", + .data = &init_net.ipv4.sysctl_tcp_fastopen_blackhole_timeout, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_tfo_blackhole_detect_timeout, + .extra1 = &zero, + }, #ifdef CONFIG_IP_ROUTE_MULTIPATH { .procname = "fib_multipath_use_neigh", diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 4eae44ac3cb0..de470e7e586f 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -422,25 +422,16 @@ EXPORT_SYMBOL(tcp_fastopen_defer_connect); * TFO connection with data exchanges. */ -/* Default to 1hr */ -unsigned int sysctl_tcp_fastopen_blackhole_timeout __read_mostly = 60 * 60; -static atomic_t tfo_active_disable_times __read_mostly = ATOMIC_INIT(0); -static unsigned long tfo_active_disable_stamp __read_mostly; - /* Disable active TFO and record current jiffies and * tfo_active_disable_times */ void tcp_fastopen_active_disable(struct sock *sk) { - atomic_inc(&tfo_active_disable_times); - tfo_active_disable_stamp = jiffies; - NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPFASTOPENBLACKHOLE); -} + struct net *net = sock_net(sk); -/* Reset tfo_active_disable_times to 0 */ -void tcp_fastopen_active_timeout_reset(void) -{ - atomic_set(&tfo_active_disable_times, 0); + atomic_inc(&net->ipv4.tfo_active_disable_times); + net->ipv4.tfo_active_disable_stamp = jiffies; + NET_INC_STATS(net, LINUX_MIB_TCPFASTOPENBLACKHOLE); } /* Calculate timeout for tfo active disable @@ -449,17 +440,18 @@ void tcp_fastopen_active_timeout_reset(void) */ bool tcp_fastopen_active_should_disable(struct sock *sk) { - int tfo_da_times = atomic_read(&tfo_active_disable_times); - int multiplier; + unsigned int tfo_bh_timeout = sock_net(sk)->ipv4.sysctl_tcp_fastopen_blackhole_timeout; + int tfo_da_times = atomic_read(&sock_net(sk)->ipv4.tfo_active_disable_times); unsigned long timeout; + int multiplier; if (!tfo_da_times) return false; /* Limit timout to max: 2^6 * initial timeout */ multiplier = 1 << min(tfo_da_times - 1, 6); - timeout = multiplier * sysctl_tcp_fastopen_blackhole_timeout * HZ; - if (time_before(jiffies, tfo_active_disable_stamp + timeout)) + timeout = multiplier * tfo_bh_timeout * HZ; + if (time_before(jiffies, sock_net(sk)->ipv4.tfo_active_disable_stamp + timeout)) return true; /* Mark check bit so we can check for successful active TFO @@ -495,10 +487,10 @@ void tcp_fastopen_active_disable_ofo_check(struct sock *sk) } } } else if (tp->syn_fastopen_ch && - atomic_read(&tfo_active_disable_times)) { + atomic_read(&sock_net(sk)->ipv4.tfo_active_disable_times)) { dst = sk_dst_get(sk); if (!(dst && dst->dev && (dst->dev->flags & IFF_LOOPBACK))) - tcp_fastopen_active_timeout_reset(); + atomic_set(&sock_net(sk)->ipv4.tfo_active_disable_times, 0); dst_release(dst); } } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 49c74c0d0d21..ad3b5bbaf942 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2474,6 +2474,8 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); + net->ipv4.sysctl_tcp_fastopen_blackhole_timeout = 60 * 60; + atomic_set(&net->ipv4.tfo_active_disable_times, 0); return 0; fail: -- cgit v1.2.3 From b80ccfe9bbcac70e66fdfaef73f0988a27f9a68c Mon Sep 17 00:00:00 2001 From: Maciej Żenczykowski Date: Tue, 26 Sep 2017 20:37:22 -0700 Subject: net-ipv6: remove unused IP6_ECN_clear() function MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This function is unused, and furthermore it is buggy since it suffers from the same issue that requires IP6_ECN_set_ce() to take a pointer to the skb so that it may (in case of CHECKSUM_COMPLETE) update skb->csum Instead of fixing it, let's just outright remove it. Tested: builds, and 'git grep IP6_ECN_clear' comes up empty Signed-off-by: Maciej Żenczykowski Signed-off-by: David S. Miller --- include/net/inet_ecn.h | 5 ----- 1 file changed, 5 deletions(-) (limited to 'include/net') diff --git a/include/net/inet_ecn.h b/include/net/inet_ecn.h index dce2d586d9ce..f5ff16d72fe6 100644 --- a/include/net/inet_ecn.h +++ b/include/net/inet_ecn.h @@ -133,11 +133,6 @@ static inline int IP6_ECN_set_ce(struct sk_buff *skb, struct ipv6hdr *iph) return 1; } -static inline void IP6_ECN_clear(struct ipv6hdr *iph) -{ - *(__be32*)iph &= ~htonl(INET_ECN_MASK << 20); -} - static inline void ipv6_copy_dscp(unsigned int dscp, struct ipv6hdr *inner) { dscp &= ~INET_ECN_MASK; -- cgit v1.2.3 From 503c1fb98ba3859c13863957c7c65c92371a9e50 Mon Sep 17 00:00:00 2001 From: Avraham Stern Date: Fri, 29 Sep 2017 14:21:49 +0200 Subject: cfg80211/nl80211: add a port authorized event Add an event that indicates that a connection is authorized (i.e. the 4 way handshake was performed by the driver). This event should be sent by the driver after sending a connect/roamed event. This is useful for networks that require 802.1X authentication. In cases that the driver supports 4 way handshake offload, but the 802.1X authentication is managed by user space, the driver needs to inform user space right after the 802.11 association was completed so user space can initialize its 802.1X state machine etc. However, it is also possible that the AP will choose to skip the 802.1X authentication (e.g. when PMKSA caching is used) and proceed with the 4 way handshake immediately. In this case the driver needs to inform user space that 802.1X authentication is no longer required (e.g. to prevent user space from disconnecting since it did not get any EAPOLs from the AP). This is also useful for roaming, in which case it is possible that the driver used the Fast Transition protocol so 802.1X is not required. Since there will now be a dedicated notification indicating that the connection is authorized, the authorized flag can be removed from the roamed event. Drivers can send the new port authorized event right after sending the roamed event to indicate the new AP is already authorized. This therefore reserves the old PORT_AUTHORIZED attribute. Signed-off-by: Avraham Stern Signed-off-by: Johannes Berg --- include/net/cfg80211.h | 21 +++++++++++++++++---- include/uapi/linux/nl80211.h | 28 ++++++++++++++++----------- net/wireless/core.h | 5 +++++ net/wireless/nl80211.c | 34 ++++++++++++++++++++++++++++++--- net/wireless/nl80211.h | 2 ++ net/wireless/sme.c | 45 +++++++++++++++++++++++++++++++++++++++++++- net/wireless/util.c | 3 +++ 7 files changed, 119 insertions(+), 19 deletions(-) (limited to 'include/net') diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index cc1996081463..8b8118a7fadb 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -5428,9 +5428,6 @@ cfg80211_connect_timeout(struct net_device *dev, const u8 *bssid, * @req_ie_len: association request IEs length * @resp_ie: association response IEs (may be %NULL) * @resp_ie_len: assoc response IEs length - * @authorized: true if the 802.1X authentication was done by the driver or is - * not needed (e.g., when Fast Transition protocol was used), false - * otherwise. Ignored for networks that don't use 802.1X authentication. */ struct cfg80211_roam_info { struct ieee80211_channel *channel; @@ -5440,7 +5437,6 @@ struct cfg80211_roam_info { size_t req_ie_len; const u8 *resp_ie; size_t resp_ie_len; - bool authorized; }; /** @@ -5464,6 +5460,23 @@ struct cfg80211_roam_info { void cfg80211_roamed(struct net_device *dev, struct cfg80211_roam_info *info, gfp_t gfp); +/** + * cfg80211_port_authorized - notify cfg80211 of successful security association + * + * @dev: network device + * @bssid: the BSSID of the AP + * @gfp: allocation flags + * + * This function should be called by a driver that supports 4 way handshake + * offload after a security association was successfully established (i.e., + * the 4 way handshake was completed successfully). The call to this function + * should be preceded with a call to cfg80211_connect_result(), + * cfg80211_connect_done(), cfg80211_connect_bss() or cfg80211_roamed() to + * indicate the 802.11 association. + */ +void cfg80211_port_authorized(struct net_device *dev, const u8 *bssid, + gfp_t gfp); + /** * cfg80211_disconnected - notify cfg80211 that connection was dropped * diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index 59ba6ca66a0d..95832ce03a44 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -569,13 +569,14 @@ * authentication/association or not receiving a response from the AP. * Non-zero %NL80211_ATTR_STATUS_CODE value is indicated in that case as * well to remain backwards compatible. - * @NL80211_CMD_ROAM: notifcation indicating the card/driver roamed by itself. - * When the driver roamed in a network that requires 802.1X authentication, - * %NL80211_ATTR_PORT_AUTHORIZED should be set if the 802.1X authentication - * was done by the driver or if roaming was done using Fast Transition - * protocol (in which case 802.1X authentication is not needed). If - * %NL80211_ATTR_PORT_AUTHORIZED is not set, user space is responsible for - * the 802.1X authentication. + * When establishing a security association, drivers that support 4 way + * handshake offload should send %NL80211_CMD_PORT_AUTHORIZED event when + * the 4 way handshake is completed successfully. + * @NL80211_CMD_ROAM: Notification indicating the card/driver roamed by itself. + * When a security association was established with the new AP (e.g. if + * the FT protocol was used for roaming or the driver completed the 4 way + * handshake), this event should be followed by an + * %NL80211_CMD_PORT_AUTHORIZED event. * @NL80211_CMD_DISCONNECT: drop a given connection; also used to notify * userspace that a connection was dropped by the AP or due to other * reasons, for this the %NL80211_ATTR_DISCONNECTED_BY_AP and @@ -982,6 +983,12 @@ * @NL80211_CMD_DEL_PMK: For offloaded 4-Way handshake, delete the previously * configured PMK for the authenticator address identified by * &NL80211_ATTR_MAC. + * @NL80211_CMD_PORT_AUTHORIZED: An event that indicates that the 4 way + * handshake was completed successfully by the driver. The BSSID is + * specified with &NL80211_ATTR_MAC. Drivers that support 4 way handshake + * offload should send this event after indicating 802.11 association with + * &NL80211_CMD_CONNECT or &NL80211_CMD_ROAM. If the 4 way handshake failed + * &NL80211_CMD_DISCONNECT should be indicated instead. * * @NL80211_CMD_MAX: highest used command number * @__NL80211_CMD_AFTER_LAST: internal use @@ -1185,6 +1192,8 @@ enum nl80211_commands { NL80211_CMD_SET_PMK, NL80211_CMD_DEL_PMK, + NL80211_CMD_PORT_AUTHORIZED, + /* add new commands above here */ /* used to define NL80211_CMD_MAX below */ @@ -2138,10 +2147,7 @@ enum nl80211_commands { * in %NL80211_CMD_CONNECT to indicate that for 802.1X authentication it * wants to use the supported offload of the 4-way handshake. * @NL80211_ATTR_PMKR0_NAME: PMK-R0 Name for offloaded FT. - * @NL80211_ATTR_PORT_AUTHORIZED: flag attribute used in %NL80211_CMD_ROAMED - * notification indicating that that 802.1X authentication was done by - * the driver or is not needed (because roaming used the Fast Transition - * protocol). + * @NL80211_ATTR_PORT_AUTHORIZED: (reserved) * * @NUM_NL80211_ATTR: total number of nl80211_attrs available * @NL80211_ATTR_MAX: highest attribute number currently defined diff --git a/net/wireless/core.h b/net/wireless/core.h index 6e809325af3b..35165f42c2a8 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -216,6 +216,7 @@ enum cfg80211_event_type { EVENT_DISCONNECTED, EVENT_IBSS_JOINED, EVENT_STOPPED, + EVENT_PORT_AUTHORIZED, }; struct cfg80211_event { @@ -235,6 +236,9 @@ struct cfg80211_event { u8 bssid[ETH_ALEN]; struct ieee80211_channel *channel; } ij; + struct { + u8 bssid[ETH_ALEN]; + } pa; }; }; @@ -385,6 +389,7 @@ int cfg80211_disconnect(struct cfg80211_registered_device *rdev, bool wextev); void __cfg80211_roamed(struct wireless_dev *wdev, struct cfg80211_roam_info *info); +void __cfg80211_port_authorized(struct wireless_dev *wdev, const u8 *bssid); int cfg80211_mgd_wext_connect(struct cfg80211_registered_device *rdev, struct wireless_dev *wdev); void cfg80211_autodisconnect_wk(struct work_struct *work); diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 1e39ba3cfd06..90e212db6889 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -13830,9 +13830,7 @@ void nl80211_send_roamed(struct cfg80211_registered_device *rdev, info->req_ie)) || (info->resp_ie && nla_put(msg, NL80211_ATTR_RESP_IE, info->resp_ie_len, - info->resp_ie)) || - (info->authorized && - nla_put_flag(msg, NL80211_ATTR_PORT_AUTHORIZED))) + info->resp_ie))) goto nla_put_failure; genlmsg_end(msg, hdr); @@ -13846,6 +13844,36 @@ void nl80211_send_roamed(struct cfg80211_registered_device *rdev, nlmsg_free(msg); } +void nl80211_send_port_authorized(struct cfg80211_registered_device *rdev, + struct net_device *netdev, const u8 *bssid) +{ + struct sk_buff *msg; + void *hdr; + + msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!msg) + return; + + hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_PORT_AUTHORIZED); + if (!hdr) { + nlmsg_free(msg); + return; + } + + if (nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, bssid)) + goto nla_put_failure; + + genlmsg_end(msg, hdr); + + genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0, + NL80211_MCGRP_MLME, GFP_KERNEL); + return; + + nla_put_failure: + genlmsg_cancel(msg, hdr); + nlmsg_free(msg); +} + void nl80211_send_disconnected(struct cfg80211_registered_device *rdev, struct net_device *netdev, u16 reason, const u8 *ie, size_t ie_len, bool from_ap) diff --git a/net/wireless/nl80211.h b/net/wireless/nl80211.h index b96933322077..bf9e772a30b9 100644 --- a/net/wireless/nl80211.h +++ b/net/wireless/nl80211.h @@ -58,6 +58,8 @@ void nl80211_send_connect_result(struct cfg80211_registered_device *rdev, void nl80211_send_roamed(struct cfg80211_registered_device *rdev, struct net_device *netdev, struct cfg80211_roam_info *info, gfp_t gfp); +void nl80211_send_port_authorized(struct cfg80211_registered_device *rdev, + struct net_device *netdev, const u8 *bssid); void nl80211_send_disconnected(struct cfg80211_registered_device *rdev, struct net_device *netdev, u16 reason, const u8 *ie, size_t ie_len, bool from_ap); diff --git a/net/wireless/sme.c b/net/wireless/sme.c index 0a49b88070d0..f38ed490e42b 100644 --- a/net/wireless/sme.c +++ b/net/wireless/sme.c @@ -960,7 +960,6 @@ void cfg80211_roamed(struct net_device *dev, struct cfg80211_roam_info *info, ev->rm.resp_ie_len = info->resp_ie_len; memcpy((void *)ev->rm.resp_ie, info->resp_ie, info->resp_ie_len); ev->rm.bss = info->bss; - ev->rm.authorized = info->authorized; spin_lock_irqsave(&wdev->event_lock, flags); list_add_tail(&ev->list, &wdev->event_list); @@ -969,6 +968,50 @@ void cfg80211_roamed(struct net_device *dev, struct cfg80211_roam_info *info, } EXPORT_SYMBOL(cfg80211_roamed); +void __cfg80211_port_authorized(struct wireless_dev *wdev, const u8 *bssid) +{ + ASSERT_WDEV_LOCK(wdev); + + if (WARN_ON(wdev->iftype != NL80211_IFTYPE_STATION)) + return; + + if (WARN_ON(!wdev->current_bss) || + WARN_ON(!ether_addr_equal(wdev->current_bss->pub.bssid, bssid))) + return; + + nl80211_send_port_authorized(wiphy_to_rdev(wdev->wiphy), wdev->netdev, + bssid); +} + +void cfg80211_port_authorized(struct net_device *dev, const u8 *bssid, + gfp_t gfp) +{ + struct wireless_dev *wdev = dev->ieee80211_ptr; + struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy); + struct cfg80211_event *ev; + unsigned long flags; + + if (WARN_ON(!bssid)) + return; + + ev = kzalloc(sizeof(*ev), gfp); + if (!ev) + return; + + ev->type = EVENT_PORT_AUTHORIZED; + memcpy(ev->pa.bssid, bssid, ETH_ALEN); + + /* + * Use the wdev event list so that if there are pending + * connected/roamed events, they will be reported first. + */ + spin_lock_irqsave(&wdev->event_lock, flags); + list_add_tail(&ev->list, &wdev->event_list); + spin_unlock_irqrestore(&wdev->event_lock, flags); + queue_work(cfg80211_wq, &rdev->event_work); +} +EXPORT_SYMBOL(cfg80211_port_authorized); + void __cfg80211_disconnected(struct net_device *dev, const u8 *ie, size_t ie_len, u16 reason, bool from_ap) { diff --git a/net/wireless/util.c b/net/wireless/util.c index 7a1fcc6ee060..ff21c314a609 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -846,6 +846,9 @@ void cfg80211_process_wdev_events(struct wireless_dev *wdev) case EVENT_STOPPED: __cfg80211_leave(wiphy_to_rdev(wdev->wiphy), wdev); break; + case EVENT_PORT_AUTHORIZED: + __cfg80211_port_authorized(wdev, ev->pa.bssid); + break; } wdev_unlock(wdev); -- cgit v1.2.3 From 32f16369e59fcc505c5ed93a6a8cad3d5636b463 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Mon, 2 Oct 2017 10:41:15 +0200 Subject: net/dst: Make skb parameter of skb{metadata_dst, tunnel_info}() const Make the skb parameter of skb_metadata_dst() and skb_tunnel_info() const as they are not modified. This is in preparation for using them in call-sites where skb is const. Signed-off-by: Simon Horman Reviewed-by: Jakub Kicinski Signed-off-by: David S. Miller --- include/net/dst_metadata.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/dst_metadata.h b/include/net/dst_metadata.h index a803129a4849..9fba2ebf6dda 100644 --- a/include/net/dst_metadata.h +++ b/include/net/dst_metadata.h @@ -24,7 +24,7 @@ struct metadata_dst { } u; }; -static inline struct metadata_dst *skb_metadata_dst(struct sk_buff *skb) +static inline struct metadata_dst *skb_metadata_dst(const struct sk_buff *skb) { struct metadata_dst *md_dst = (struct metadata_dst *) skb_dst(skb); @@ -34,7 +34,8 @@ static inline struct metadata_dst *skb_metadata_dst(struct sk_buff *skb) return NULL; } -static inline struct ip_tunnel_info *skb_tunnel_info(struct sk_buff *skb) +static inline struct ip_tunnel_info * +skb_tunnel_info(const struct sk_buff *skb) { struct metadata_dst *md_dst = skb_metadata_dst(skb); struct dst_entry *dst; -- cgit v1.2.3 From f952be79cebd49d04154781d99408867a069d375 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Tue, 3 Oct 2017 19:20:11 -0300 Subject: sctp: introduce struct sctp_stream_out_ext With the stream schedulers, sctp_stream_out will become too big to be allocated by kmalloc and as we need to allocate with BH disabled, we cannot use __vmalloc in sctp_stream_init(). This patch moves out the stats from sctp_stream_out to sctp_stream_out_ext, which will be allocated only when the application tries to sendmsg something on it. Just the introduction of sctp_stream_out_ext would already fix the issue described above by splitting the allocation in two. Moving the stats to it also reduces the pressure on the allocator as we will ask for less memory atomically when creating the socket and we will use GFP_KERNEL later. Then, for stream schedulers, we will just use sctp_stream_out_ext. Tested-by: Xin Long Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 10 ++++++++-- net/sctp/chunk.c | 6 +++--- net/sctp/outqueue.c | 4 ++-- net/sctp/socket.c | 27 +++++++++++++++++++++------ net/sctp/stream.c | 16 ++++++++++++++++ 5 files changed, 50 insertions(+), 13 deletions(-) (limited to 'include/net') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 0477945de1a3..9b2b30b3ba4d 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -84,6 +84,7 @@ struct sctp_ulpq; struct sctp_ep_common; struct crypto_shash; struct sctp_stream; +struct sctp_stream_out; #include @@ -380,6 +381,7 @@ struct sctp_sender_hb_info { int sctp_stream_init(struct sctp_stream *stream, __u16 outcnt, __u16 incnt, gfp_t gfp); +int sctp_stream_init_ext(struct sctp_stream *stream, __u16 sid); void sctp_stream_free(struct sctp_stream *stream); void sctp_stream_clear(struct sctp_stream *stream); void sctp_stream_update(struct sctp_stream *stream, struct sctp_stream *new); @@ -1315,11 +1317,15 @@ struct sctp_inithdr_host { __u32 initial_tsn; }; +struct sctp_stream_out_ext { + __u64 abandoned_unsent[SCTP_PR_INDEX(MAX) + 1]; + __u64 abandoned_sent[SCTP_PR_INDEX(MAX) + 1]; +}; + struct sctp_stream_out { __u16 ssn; __u8 state; - __u64 abandoned_unsent[SCTP_PR_INDEX(MAX) + 1]; - __u64 abandoned_sent[SCTP_PR_INDEX(MAX) + 1]; + struct sctp_stream_out_ext *ext; }; struct sctp_stream_in { diff --git a/net/sctp/chunk.c b/net/sctp/chunk.c index 3afac275ee82..7b261afc47b9 100644 --- a/net/sctp/chunk.c +++ b/net/sctp/chunk.c @@ -311,10 +311,10 @@ int sctp_chunk_abandoned(struct sctp_chunk *chunk) if (chunk->sent_count) { chunk->asoc->abandoned_sent[SCTP_PR_INDEX(TTL)]++; - streamout->abandoned_sent[SCTP_PR_INDEX(TTL)]++; + streamout->ext->abandoned_sent[SCTP_PR_INDEX(TTL)]++; } else { chunk->asoc->abandoned_unsent[SCTP_PR_INDEX(TTL)]++; - streamout->abandoned_unsent[SCTP_PR_INDEX(TTL)]++; + streamout->ext->abandoned_unsent[SCTP_PR_INDEX(TTL)]++; } return 1; } else if (SCTP_PR_RTX_ENABLED(chunk->sinfo.sinfo_flags) && @@ -323,7 +323,7 @@ int sctp_chunk_abandoned(struct sctp_chunk *chunk) &chunk->asoc->stream.out[chunk->sinfo.sinfo_stream]; chunk->asoc->abandoned_sent[SCTP_PR_INDEX(RTX)]++; - streamout->abandoned_sent[SCTP_PR_INDEX(RTX)]++; + streamout->ext->abandoned_sent[SCTP_PR_INDEX(RTX)]++; return 1; } else if (!SCTP_PR_POLICY(chunk->sinfo.sinfo_flags) && chunk->msg->expires_at && diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c index 2966ff400755..746b07b7937d 100644 --- a/net/sctp/outqueue.c +++ b/net/sctp/outqueue.c @@ -366,7 +366,7 @@ static int sctp_prsctp_prune_sent(struct sctp_association *asoc, streamout = &asoc->stream.out[chk->sinfo.sinfo_stream]; asoc->sent_cnt_removable--; asoc->abandoned_sent[SCTP_PR_INDEX(PRIO)]++; - streamout->abandoned_sent[SCTP_PR_INDEX(PRIO)]++; + streamout->ext->abandoned_sent[SCTP_PR_INDEX(PRIO)]++; if (!chk->tsn_gap_acked) { if (chk->transport) @@ -404,7 +404,7 @@ static int sctp_prsctp_prune_unsent(struct sctp_association *asoc, struct sctp_stream_out *streamout = &asoc->stream.out[chk->sinfo.sinfo_stream]; - streamout->abandoned_unsent[SCTP_PR_INDEX(PRIO)]++; + streamout->ext->abandoned_unsent[SCTP_PR_INDEX(PRIO)]++; } msg_len -= SCTP_DATA_SNDSIZE(chk) + diff --git a/net/sctp/socket.c b/net/sctp/socket.c index d4730ada7f32..d207734326b0 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -1927,6 +1927,13 @@ static int sctp_sendmsg(struct sock *sk, struct msghdr *msg, size_t msg_len) goto out_free; } + /* Allocate sctp_stream_out_ext if not already done */ + if (unlikely(!asoc->stream.out[sinfo->sinfo_stream].ext)) { + err = sctp_stream_init_ext(&asoc->stream, sinfo->sinfo_stream); + if (err) + goto out_free; + } + if (sctp_wspace(asoc) < msg_len) sctp_prsctp_prune(asoc, sinfo, msg_len - sctp_wspace(asoc)); @@ -6645,7 +6652,7 @@ static int sctp_getsockopt_pr_streamstatus(struct sock *sk, int len, char __user *optval, int __user *optlen) { - struct sctp_stream_out *streamout; + struct sctp_stream_out_ext *streamoute; struct sctp_association *asoc; struct sctp_prstatus params; int retval = -EINVAL; @@ -6668,21 +6675,29 @@ static int sctp_getsockopt_pr_streamstatus(struct sock *sk, int len, if (!asoc || params.sprstat_sid >= asoc->stream.outcnt) goto out; - streamout = &asoc->stream.out[params.sprstat_sid]; + streamoute = asoc->stream.out[params.sprstat_sid].ext; + if (!streamoute) { + /* Not allocated yet, means all stats are 0 */ + params.sprstat_abandoned_unsent = 0; + params.sprstat_abandoned_sent = 0; + retval = 0; + goto out; + } + if (policy == SCTP_PR_SCTP_NONE) { params.sprstat_abandoned_unsent = 0; params.sprstat_abandoned_sent = 0; for (policy = 0; policy <= SCTP_PR_INDEX(MAX); policy++) { params.sprstat_abandoned_unsent += - streamout->abandoned_unsent[policy]; + streamoute->abandoned_unsent[policy]; params.sprstat_abandoned_sent += - streamout->abandoned_sent[policy]; + streamoute->abandoned_sent[policy]; } } else { params.sprstat_abandoned_unsent = - streamout->abandoned_unsent[__SCTP_PR_INDEX(policy)]; + streamoute->abandoned_unsent[__SCTP_PR_INDEX(policy)]; params.sprstat_abandoned_sent = - streamout->abandoned_sent[__SCTP_PR_INDEX(policy)]; + streamoute->abandoned_sent[__SCTP_PR_INDEX(policy)]; } if (put_user(len, optlen) || copy_to_user(optval, ¶ms, len)) { diff --git a/net/sctp/stream.c b/net/sctp/stream.c index 952437d656cc..055ca25bbc91 100644 --- a/net/sctp/stream.c +++ b/net/sctp/stream.c @@ -121,8 +121,24 @@ in: return 0; } +int sctp_stream_init_ext(struct sctp_stream *stream, __u16 sid) +{ + struct sctp_stream_out_ext *soute; + + soute = kzalloc(sizeof(*soute), GFP_KERNEL); + if (!soute) + return -ENOMEM; + stream->out[sid].ext = soute; + + return 0; +} + void sctp_stream_free(struct sctp_stream *stream) { + int i; + + for (i = 0; i < stream->outcnt; i++) + kfree(stream->out[i].ext); kfree(stream->out); kfree(stream->in); } -- cgit v1.2.3 From 2fc019f790312e703efa1a44204c586112a430dc Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Tue, 3 Oct 2017 19:20:12 -0300 Subject: sctp: introduce sctp_chunk_stream_no Add a helper to fetch the stream number from a given chunk. Tested-by: Xin Long Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/net') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 9b2b30b3ba4d..c48f7999fe9b 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -642,6 +642,11 @@ void sctp_init_addrs(struct sctp_chunk *, union sctp_addr *, union sctp_addr *); const union sctp_addr *sctp_source(const struct sctp_chunk *chunk); +static inline __u16 sctp_chunk_stream_no(struct sctp_chunk *ch) +{ + return ntohs(ch->subh.data_hdr->stream); +} + enum { SCTP_ADDR_NEW, /* new address added to assoc/ep */ SCTP_ADDR_SRC, /* address can be used as source */ -- cgit v1.2.3 From 5bbbbe32a43199c2b9ea5ea66fab6241c64beb51 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Tue, 3 Oct 2017 19:20:13 -0300 Subject: sctp: introduce stream scheduler foundations This patch introduces the hooks necessary to do stream scheduling, as per RFC Draft ndata. It also introduces the first scheduler, which is what we do today but now factored out: first come first served (FCFS). With stream scheduling now we have to track which chunk was enqueued on which stream and be able to select another other than the in front of the main outqueue. So we introduce a list on sctp_stream_out_ext structure for this purpose. We reuse sctp_chunk->transmitted_list space for the list above, as the chunk cannot belong to the two lists at the same time. By using the union in there, we can have distinct names for these moments. sctp_sched_ops are the operations expected to be implemented by each scheduler. The dequeueing is a bit particular to this implementation but it is to match how we dequeue packets today. We first dequeue and then check if it fits the packet and if not, we requeue it at head. Thus why we don't have a peek operation but have dequeue_done instead, which is called once the chunk can be safely considered as transmitted. The check removed from sctp_outq_flush is now performed by sctp_stream_outq_migrate, which is only called during assoc setup. (sctp_sendmsg() also checks for it) The only operation that is foreseen but not yet added here is a way to signalize that a new packet is starting or that the packet is done, for round robin scheduler per packet, but is intentionally left to the patch that actually implements it. Support for I-DATA chunks, also described in this RFC, with user message interleaving is straightforward as it just requires the schedulers to probe for the feature and ignore datamsg boundaries when dequeueing. See-also: https://tools.ietf.org/html/draft-ietf-tsvwg-sctp-ndata-13 Tested-by: Xin Long Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sctp/stream_sched.h | 72 +++++++++++ include/net/sctp/structs.h | 15 ++- include/uapi/linux/sctp.h | 6 + net/sctp/Makefile | 2 +- net/sctp/outqueue.c | 59 +++++---- net/sctp/sm_sideeffect.c | 3 + net/sctp/stream.c | 88 +++++++++++-- net/sctp/stream_sched.c | 270 ++++++++++++++++++++++++++++++++++++++++ 8 files changed, 477 insertions(+), 38 deletions(-) create mode 100644 include/net/sctp/stream_sched.h create mode 100644 net/sctp/stream_sched.c (limited to 'include/net') diff --git a/include/net/sctp/stream_sched.h b/include/net/sctp/stream_sched.h new file mode 100644 index 000000000000..c676550a4c7d --- /dev/null +++ b/include/net/sctp/stream_sched.h @@ -0,0 +1,72 @@ +/* SCTP kernel implementation + * (C) Copyright Red Hat Inc. 2017 + * + * These are definitions used by the stream schedulers, defined in RFC + * draft ndata (https://tools.ietf.org/html/draft-ietf-tsvwg-sctp-ndata-11) + * + * This SCTP implementation is free software; + * you can redistribute it and/or modify it under the terms of + * the GNU General Public License as published by + * the Free Software Foundation; either version 2, or (at your option) + * any later version. + * + * This SCTP implementation 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 GNU CC; see the file COPYING. If not, see + * . + * + * Please send any bug reports or fixes you make to the + * email addresses: + * lksctp developers + * + * Written or modified by: + * Marcelo Ricardo Leitner + */ + +#ifndef __sctp_stream_sched_h__ +#define __sctp_stream_sched_h__ + +struct sctp_sched_ops { + /* Property handling for a given stream */ + int (*set)(struct sctp_stream *stream, __u16 sid, __u16 value, + gfp_t gfp); + int (*get)(struct sctp_stream *stream, __u16 sid, __u16 *value); + + /* Init the specific scheduler */ + int (*init)(struct sctp_stream *stream); + /* Init a stream */ + int (*init_sid)(struct sctp_stream *stream, __u16 sid, gfp_t gfp); + /* Frees the entire thing */ + void (*free)(struct sctp_stream *stream); + + /* Enqueue a chunk */ + void (*enqueue)(struct sctp_outq *q, struct sctp_datamsg *msg); + /* Dequeue a chunk */ + struct sctp_chunk *(*dequeue)(struct sctp_outq *q); + /* Called only if the chunk fit the packet */ + void (*dequeue_done)(struct sctp_outq *q, struct sctp_chunk *chunk); + /* Sched all chunks already enqueued */ + void (*sched_all)(struct sctp_stream *steam); + /* Unched all chunks already enqueued */ + void (*unsched_all)(struct sctp_stream *steam); +}; + +int sctp_sched_set_sched(struct sctp_association *asoc, + enum sctp_sched_type sched); +int sctp_sched_get_sched(struct sctp_association *asoc); +int sctp_sched_set_value(struct sctp_association *asoc, __u16 sid, + __u16 value, gfp_t gfp); +int sctp_sched_get_value(struct sctp_association *asoc, __u16 sid, + __u16 *value); +void sctp_sched_dequeue_done(struct sctp_outq *q, struct sctp_chunk *ch); + +void sctp_sched_dequeue_common(struct sctp_outq *q, struct sctp_chunk *ch); +int sctp_sched_init_sid(struct sctp_stream *stream, __u16 sid, gfp_t gfp); +struct sctp_sched_ops *sctp_sched_ops_from_stream(struct sctp_stream *stream); + +#endif /* __sctp_stream_sched_h__ */ diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index c48f7999fe9b..3c22a30fd71b 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -84,7 +84,6 @@ struct sctp_ulpq; struct sctp_ep_common; struct crypto_shash; struct sctp_stream; -struct sctp_stream_out; #include @@ -531,8 +530,12 @@ struct sctp_chunk { /* How many times this chunk have been sent, for prsctp RTX policy */ int sent_count; - /* This is our link to the per-transport transmitted list. */ - struct list_head transmitted_list; + union { + /* This is our link to the per-transport transmitted list. */ + struct list_head transmitted_list; + /* List in specific stream outq */ + struct list_head stream_list; + }; /* This field is used by chunks that hold fragmented data. * For the first fragment this is the list that holds the rest of @@ -1019,6 +1022,9 @@ struct sctp_outq { /* Data pending that has never been transmitted. */ struct list_head out_chunk_list; + /* Stream scheduler being used */ + struct sctp_sched_ops *sched; + unsigned int out_qlen; /* Total length of queued data chunks. */ /* Error of send failed, may used in SCTP_SEND_FAILED event. */ @@ -1325,6 +1331,7 @@ struct sctp_inithdr_host { struct sctp_stream_out_ext { __u64 abandoned_unsent[SCTP_PR_INDEX(MAX) + 1]; __u64 abandoned_sent[SCTP_PR_INDEX(MAX) + 1]; + struct list_head outq; /* chunks enqueued by this stream */ }; struct sctp_stream_out { @@ -1342,6 +1349,8 @@ struct sctp_stream { struct sctp_stream_in *in; __u16 outcnt; __u16 incnt; + /* Current stream being sent, if any */ + struct sctp_stream_out *out_curr; }; #define SCTP_STREAM_CLOSED 0x00 diff --git a/include/uapi/linux/sctp.h b/include/uapi/linux/sctp.h index 6217ff8500a1..4487e7625ddb 100644 --- a/include/uapi/linux/sctp.h +++ b/include/uapi/linux/sctp.h @@ -1088,4 +1088,10 @@ struct sctp_add_streams { uint16_t sas_outstrms; }; +/* SCTP Stream schedulers */ +enum sctp_sched_type { + SCTP_SS_FCFS, + SCTP_SS_MAX = SCTP_SS_FCFS +}; + #endif /* _UAPI_SCTP_H */ diff --git a/net/sctp/Makefile b/net/sctp/Makefile index 70f1b570bab9..0f6e6d1d69fd 100644 --- a/net/sctp/Makefile +++ b/net/sctp/Makefile @@ -12,7 +12,7 @@ sctp-y := sm_statetable.o sm_statefuns.o sm_sideeffect.o \ inqueue.o outqueue.o ulpqueue.o \ tsnmap.o bind_addr.o socket.o primitive.o \ output.o input.o debug.o stream.o auth.o \ - offload.o + offload.o stream_sched.o sctp_probe-y := probe.o diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c index 746b07b7937d..4db012aa25f7 100644 --- a/net/sctp/outqueue.c +++ b/net/sctp/outqueue.c @@ -50,6 +50,7 @@ #include #include +#include /* Declare internal functions here. */ static int sctp_acked(struct sctp_sackhdr *sack, __u32 tsn); @@ -72,32 +73,38 @@ static void sctp_outq_flush(struct sctp_outq *q, int rtx_timeout, gfp_t gfp); /* Add data to the front of the queue. */ static inline void sctp_outq_head_data(struct sctp_outq *q, - struct sctp_chunk *ch) + struct sctp_chunk *ch) { + struct sctp_stream_out_ext *oute; + __u16 stream; + list_add(&ch->list, &q->out_chunk_list); q->out_qlen += ch->skb->len; + + stream = sctp_chunk_stream_no(ch); + oute = q->asoc->stream.out[stream].ext; + list_add(&ch->stream_list, &oute->outq); } /* Take data from the front of the queue. */ static inline struct sctp_chunk *sctp_outq_dequeue_data(struct sctp_outq *q) { - struct sctp_chunk *ch = NULL; - - if (!list_empty(&q->out_chunk_list)) { - struct list_head *entry = q->out_chunk_list.next; - - ch = list_entry(entry, struct sctp_chunk, list); - list_del_init(entry); - q->out_qlen -= ch->skb->len; - } - return ch; + return q->sched->dequeue(q); } + /* Add data chunk to the end of the queue. */ static inline void sctp_outq_tail_data(struct sctp_outq *q, struct sctp_chunk *ch) { + struct sctp_stream_out_ext *oute; + __u16 stream; + list_add_tail(&ch->list, &q->out_chunk_list); q->out_qlen += ch->skb->len; + + stream = sctp_chunk_stream_no(ch); + oute = q->asoc->stream.out[stream].ext; + list_add_tail(&ch->stream_list, &oute->outq); } /* @@ -207,6 +214,7 @@ void sctp_outq_init(struct sctp_association *asoc, struct sctp_outq *q) INIT_LIST_HEAD(&q->retransmit); INIT_LIST_HEAD(&q->sacked); INIT_LIST_HEAD(&q->abandoned); + sctp_sched_set_sched(asoc, SCTP_SS_FCFS); } /* Free the outqueue structure and any related pending chunks. @@ -258,6 +266,7 @@ static void __sctp_outq_teardown(struct sctp_outq *q) /* Throw away any leftover data chunks. */ while ((chunk = sctp_outq_dequeue_data(q)) != NULL) { + sctp_sched_dequeue_done(q, chunk); /* Mark as send failure. */ sctp_chunk_fail(chunk, q->error); @@ -391,13 +400,14 @@ static int sctp_prsctp_prune_unsent(struct sctp_association *asoc, struct sctp_outq *q = &asoc->outqueue; struct sctp_chunk *chk, *temp; + q->sched->unsched_all(&asoc->stream); + list_for_each_entry_safe(chk, temp, &q->out_chunk_list, list) { if (!SCTP_PR_PRIO_ENABLED(chk->sinfo.sinfo_flags) || chk->sinfo.sinfo_timetolive <= sinfo->sinfo_timetolive) continue; - list_del_init(&chk->list); - q->out_qlen -= chk->skb->len; + sctp_sched_dequeue_common(q, chk); asoc->sent_cnt_removable--; asoc->abandoned_unsent[SCTP_PR_INDEX(PRIO)]++; if (chk->sinfo.sinfo_stream < asoc->stream.outcnt) { @@ -415,6 +425,8 @@ static int sctp_prsctp_prune_unsent(struct sctp_association *asoc, break; } + q->sched->sched_all(&asoc->stream); + return msg_len; } @@ -1033,22 +1045,9 @@ static void sctp_outq_flush(struct sctp_outq *q, int rtx_timeout, gfp_t gfp) while ((chunk = sctp_outq_dequeue_data(q)) != NULL) { __u32 sid = ntohs(chunk->subh.data_hdr->stream); - /* RFC 2960 6.5 Every DATA chunk MUST carry a valid - * stream identifier. - */ - if (chunk->sinfo.sinfo_stream >= asoc->stream.outcnt) { - - /* Mark as failed send. */ - sctp_chunk_fail(chunk, SCTP_ERROR_INV_STRM); - if (asoc->peer.prsctp_capable && - SCTP_PR_PRIO_ENABLED(chunk->sinfo.sinfo_flags)) - asoc->sent_cnt_removable--; - sctp_chunk_free(chunk); - continue; - } - /* Has this chunk expired? */ if (sctp_chunk_abandoned(chunk)) { + sctp_sched_dequeue_done(q, chunk); sctp_chunk_fail(chunk, 0); sctp_chunk_free(chunk); continue; @@ -1070,6 +1069,7 @@ static void sctp_outq_flush(struct sctp_outq *q, int rtx_timeout, gfp_t gfp) new_transport = asoc->peer.active_path; if (new_transport->state == SCTP_UNCONFIRMED) { WARN_ONCE(1, "Attempt to send packet on unconfirmed path."); + sctp_sched_dequeue_done(q, chunk); sctp_chunk_fail(chunk, 0); sctp_chunk_free(chunk); continue; @@ -1133,6 +1133,11 @@ static void sctp_outq_flush(struct sctp_outq *q, int rtx_timeout, gfp_t gfp) else asoc->stats.oodchunks++; + /* Only now it's safe to consider this + * chunk as sent, sched-wise. + */ + sctp_sched_dequeue_done(q, chunk); + break; default: diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index e6a2974e020e..402bfbb888cd 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -50,6 +50,7 @@ #include #include #include +#include static int sctp_cmd_interpreter(enum sctp_event event_type, union sctp_subtype subtype, @@ -1089,6 +1090,8 @@ static void sctp_cmd_send_msg(struct sctp_association *asoc, list_for_each_entry(chunk, &msg->chunks, frag_list) sctp_outq_tail(&asoc->outqueue, chunk, gfp); + + asoc->outqueue.sched->enqueue(&asoc->outqueue, msg); } diff --git a/net/sctp/stream.c b/net/sctp/stream.c index 055ca25bbc91..5ea33a2c453b 100644 --- a/net/sctp/stream.c +++ b/net/sctp/stream.c @@ -32,8 +32,61 @@ * Xin Long */ +#include #include #include +#include + +/* Migrates chunks from stream queues to new stream queues if needed, + * but not across associations. Also, removes those chunks to streams + * higher than the new max. + */ +static void sctp_stream_outq_migrate(struct sctp_stream *stream, + struct sctp_stream *new, __u16 outcnt) +{ + struct sctp_association *asoc; + struct sctp_chunk *ch, *temp; + struct sctp_outq *outq; + int i; + + asoc = container_of(stream, struct sctp_association, stream); + outq = &asoc->outqueue; + + list_for_each_entry_safe(ch, temp, &outq->out_chunk_list, list) { + __u16 sid = sctp_chunk_stream_no(ch); + + if (sid < outcnt) + continue; + + sctp_sched_dequeue_common(outq, ch); + /* No need to call dequeue_done here because + * the chunks are not scheduled by now. + */ + + /* Mark as failed send. */ + sctp_chunk_fail(ch, SCTP_ERROR_INV_STRM); + if (asoc->peer.prsctp_capable && + SCTP_PR_PRIO_ENABLED(ch->sinfo.sinfo_flags)) + asoc->sent_cnt_removable--; + + sctp_chunk_free(ch); + } + + if (new) { + /* Here we actually move the old ext stuff into the new + * buffer, because we want to keep it. Then + * sctp_stream_update will swap ->out pointers. + */ + for (i = 0; i < outcnt; i++) { + kfree(new->out[i].ext); + new->out[i].ext = stream->out[i].ext; + stream->out[i].ext = NULL; + } + } + + for (i = outcnt; i < stream->outcnt; i++) + kfree(stream->out[i].ext); +} static int sctp_stream_alloc_out(struct sctp_stream *stream, __u16 outcnt, gfp_t gfp) @@ -87,7 +140,8 @@ static int sctp_stream_alloc_in(struct sctp_stream *stream, __u16 incnt, int sctp_stream_init(struct sctp_stream *stream, __u16 outcnt, __u16 incnt, gfp_t gfp) { - int i; + struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream); + int i, ret = 0; gfp |= __GFP_NOWARN; @@ -97,6 +151,11 @@ int sctp_stream_init(struct sctp_stream *stream, __u16 outcnt, __u16 incnt, if (outcnt == stream->outcnt) goto in; + /* Filter out chunks queued on streams that won't exist anymore */ + sched->unsched_all(stream); + sctp_stream_outq_migrate(stream, NULL, outcnt); + sched->sched_all(stream); + i = sctp_stream_alloc_out(stream, outcnt, gfp); if (i) return i; @@ -105,20 +164,27 @@ int sctp_stream_init(struct sctp_stream *stream, __u16 outcnt, __u16 incnt, for (i = 0; i < stream->outcnt; i++) stream->out[i].state = SCTP_STREAM_OPEN; + sched->init(stream); + in: if (!incnt) - return 0; + goto out; i = sctp_stream_alloc_in(stream, incnt, gfp); if (i) { - kfree(stream->out); - stream->out = NULL; - return -ENOMEM; + ret = -ENOMEM; + goto free; } stream->incnt = incnt; + goto out; - return 0; +free: + sched->free(stream); + kfree(stream->out); + stream->out = NULL; +out: + return ret; } int sctp_stream_init_ext(struct sctp_stream *stream, __u16 sid) @@ -130,13 +196,15 @@ int sctp_stream_init_ext(struct sctp_stream *stream, __u16 sid) return -ENOMEM; stream->out[sid].ext = soute; - return 0; + return sctp_sched_init_sid(stream, sid, GFP_KERNEL); } void sctp_stream_free(struct sctp_stream *stream) { + struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream); int i; + sched->free(stream); for (i = 0; i < stream->outcnt; i++) kfree(stream->out[i].ext); kfree(stream->out); @@ -156,6 +224,10 @@ void sctp_stream_clear(struct sctp_stream *stream) void sctp_stream_update(struct sctp_stream *stream, struct sctp_stream *new) { + struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream); + + sched->unsched_all(stream); + sctp_stream_outq_migrate(stream, new, new->outcnt); sctp_stream_free(stream); stream->out = new->out; @@ -163,6 +235,8 @@ void sctp_stream_update(struct sctp_stream *stream, struct sctp_stream *new) stream->outcnt = new->outcnt; stream->incnt = new->incnt; + sched->sched_all(stream); + new->out = NULL; new->in = NULL; } diff --git a/net/sctp/stream_sched.c b/net/sctp/stream_sched.c new file mode 100644 index 000000000000..40a9a9de2b98 --- /dev/null +++ b/net/sctp/stream_sched.c @@ -0,0 +1,270 @@ +/* SCTP kernel implementation + * (C) Copyright Red Hat Inc. 2017 + * + * This file is part of the SCTP kernel implementation + * + * These functions manipulate sctp stream queue/scheduling. + * + * This SCTP implementation is free software; + * you can redistribute it and/or modify it under the terms of + * the GNU General Public License as published by + * the Free Software Foundation; either version 2, or (at your option) + * any later version. + * + * This SCTP implementation 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 GNU CC; see the file COPYING. If not, see + * . + * + * Please send any bug reports or fixes you make to the + * email addresched(es): + * lksctp developers + * + * Written or modified by: + * Marcelo Ricardo Leitner + */ + +#include +#include +#include +#include + +/* First Come First Serve (a.k.a. FIFO) + * RFC DRAFT ndata Section 3.1 + */ +static int sctp_sched_fcfs_set(struct sctp_stream *stream, __u16 sid, + __u16 value, gfp_t gfp) +{ + return 0; +} + +static int sctp_sched_fcfs_get(struct sctp_stream *stream, __u16 sid, + __u16 *value) +{ + *value = 0; + return 0; +} + +static int sctp_sched_fcfs_init(struct sctp_stream *stream) +{ + return 0; +} + +static int sctp_sched_fcfs_init_sid(struct sctp_stream *stream, __u16 sid, + gfp_t gfp) +{ + return 0; +} + +static void sctp_sched_fcfs_free(struct sctp_stream *stream) +{ +} + +static void sctp_sched_fcfs_enqueue(struct sctp_outq *q, + struct sctp_datamsg *msg) +{ +} + +static struct sctp_chunk *sctp_sched_fcfs_dequeue(struct sctp_outq *q) +{ + struct sctp_stream *stream = &q->asoc->stream; + struct sctp_chunk *ch = NULL; + struct list_head *entry; + + if (list_empty(&q->out_chunk_list)) + goto out; + + if (stream->out_curr) { + ch = list_entry(stream->out_curr->ext->outq.next, + struct sctp_chunk, stream_list); + } else { + entry = q->out_chunk_list.next; + ch = list_entry(entry, struct sctp_chunk, list); + } + + sctp_sched_dequeue_common(q, ch); + +out: + return ch; +} + +static void sctp_sched_fcfs_dequeue_done(struct sctp_outq *q, + struct sctp_chunk *chunk) +{ +} + +static void sctp_sched_fcfs_sched_all(struct sctp_stream *stream) +{ +} + +static void sctp_sched_fcfs_unsched_all(struct sctp_stream *stream) +{ +} + +static struct sctp_sched_ops sctp_sched_fcfs = { + .set = sctp_sched_fcfs_set, + .get = sctp_sched_fcfs_get, + .init = sctp_sched_fcfs_init, + .init_sid = sctp_sched_fcfs_init_sid, + .free = sctp_sched_fcfs_free, + .enqueue = sctp_sched_fcfs_enqueue, + .dequeue = sctp_sched_fcfs_dequeue, + .dequeue_done = sctp_sched_fcfs_dequeue_done, + .sched_all = sctp_sched_fcfs_sched_all, + .unsched_all = sctp_sched_fcfs_unsched_all, +}; + +/* API to other parts of the stack */ + +struct sctp_sched_ops *sctp_sched_ops[] = { + &sctp_sched_fcfs, +}; + +int sctp_sched_set_sched(struct sctp_association *asoc, + enum sctp_sched_type sched) +{ + struct sctp_sched_ops *n = sctp_sched_ops[sched]; + struct sctp_sched_ops *old = asoc->outqueue.sched; + struct sctp_datamsg *msg = NULL; + struct sctp_chunk *ch; + int i, ret = 0; + + if (old == n) + return ret; + + if (sched > SCTP_SS_MAX) + return -EINVAL; + + if (old) { + old->free(&asoc->stream); + + /* Give the next scheduler a clean slate. */ + for (i = 0; i < asoc->stream.outcnt; i++) { + void *p = asoc->stream.out[i].ext; + + if (!p) + continue; + + p += offsetofend(struct sctp_stream_out_ext, outq); + memset(p, 0, sizeof(struct sctp_stream_out_ext) - + offsetofend(struct sctp_stream_out_ext, outq)); + } + } + + asoc->outqueue.sched = n; + n->init(&asoc->stream); + for (i = 0; i < asoc->stream.outcnt; i++) { + if (!asoc->stream.out[i].ext) + continue; + + ret = n->init_sid(&asoc->stream, i, GFP_KERNEL); + if (ret) + goto err; + } + + /* We have to requeue all chunks already queued. */ + list_for_each_entry(ch, &asoc->outqueue.out_chunk_list, list) { + if (ch->msg == msg) + continue; + msg = ch->msg; + n->enqueue(&asoc->outqueue, msg); + } + + return ret; + +err: + n->free(&asoc->stream); + asoc->outqueue.sched = &sctp_sched_fcfs; /* Always safe */ + + return ret; +} + +int sctp_sched_get_sched(struct sctp_association *asoc) +{ + int i; + + for (i = 0; i <= SCTP_SS_MAX; i++) + if (asoc->outqueue.sched == sctp_sched_ops[i]) + return i; + + return 0; +} + +int sctp_sched_set_value(struct sctp_association *asoc, __u16 sid, + __u16 value, gfp_t gfp) +{ + if (sid >= asoc->stream.outcnt) + return -EINVAL; + + if (!asoc->stream.out[sid].ext) { + int ret; + + ret = sctp_stream_init_ext(&asoc->stream, sid); + if (ret) + return ret; + } + + return asoc->outqueue.sched->set(&asoc->stream, sid, value, gfp); +} + +int sctp_sched_get_value(struct sctp_association *asoc, __u16 sid, + __u16 *value) +{ + if (sid >= asoc->stream.outcnt) + return -EINVAL; + + if (!asoc->stream.out[sid].ext) + return 0; + + return asoc->outqueue.sched->get(&asoc->stream, sid, value); +} + +void sctp_sched_dequeue_done(struct sctp_outq *q, struct sctp_chunk *ch) +{ + if (!list_is_last(&ch->frag_list, &ch->msg->chunks)) { + struct sctp_stream_out *sout; + __u16 sid; + + /* datamsg is not finish, so save it as current one, + * in case application switch scheduler or a higher + * priority stream comes in. + */ + sid = sctp_chunk_stream_no(ch); + sout = &q->asoc->stream.out[sid]; + q->asoc->stream.out_curr = sout; + return; + } + + q->asoc->stream.out_curr = NULL; + q->sched->dequeue_done(q, ch); +} + +/* Auxiliary functions for the schedulers */ +void sctp_sched_dequeue_common(struct sctp_outq *q, struct sctp_chunk *ch) +{ + list_del_init(&ch->list); + list_del_init(&ch->stream_list); + q->out_qlen -= ch->skb->len; +} + +int sctp_sched_init_sid(struct sctp_stream *stream, __u16 sid, gfp_t gfp) +{ + struct sctp_sched_ops *sched = sctp_sched_ops_from_stream(stream); + + INIT_LIST_HEAD(&stream->out[sid].ext->outq); + return sched->init_sid(stream, sid, gfp); +} + +struct sctp_sched_ops *sctp_sched_ops_from_stream(struct sctp_stream *stream) +{ + struct sctp_association *asoc; + + asoc = container_of(stream, struct sctp_association, stream); + + return asoc->outqueue.sched; +} -- cgit v1.2.3 From 637784ade221a3c8a7ecd0f583eddd95d6276b9a Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Tue, 3 Oct 2017 19:20:16 -0300 Subject: sctp: introduce priority based stream scheduler This patch introduces RFC Draft ndata section 3.4 Priority Based Scheduler (SCTP_SS_PRIO). It works by having a struct sctp_stream_priority for each priority configured. This struct is then enlisted on a queue ordered per priority if, and only if, there is a stream with data queued, so that dequeueing is very straightforward: either finish current datamsg or simply dequeue from the highest priority queued, which is the next stream pointed, and that's it. If there are multiple streams assigned with the same priority and with data queued, it will do round robin amongst them while respecting datamsgs boundaries (when not using idata chunks), to be reasonably fair. We intentionally don't maintain a list of priorities nor a list of all streams with the same priority to save memory. The first would mean at least 2 other pointers per priority (which, for 1000 priorities, that can mean 16kB) and the second would also mean 2 other pointers but per stream. As SCTP supports up to 65535 streams on a given asoc, that's 1MB. This impacts when giving a priority to some stream, as we have to find out if the new priority is already being used and if we can free the old one, and also when tearing down. The new fields in struct sctp_stream_out_ext and sctp_stream are added under a union because that memory is to be shared with other schedulers. See-also: https://tools.ietf.org/html/draft-ietf-tsvwg-sctp-ndata-13 Tested-by: Xin Long Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 24 +++ include/uapi/linux/sctp.h | 3 +- net/sctp/Makefile | 2 +- net/sctp/stream_sched.c | 3 + net/sctp/stream_sched_prio.c | 347 +++++++++++++++++++++++++++++++++++++++++++ 5 files changed, 377 insertions(+), 2 deletions(-) create mode 100644 net/sctp/stream_sched_prio.c (limited to 'include/net') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 3c22a30fd71b..40eb8d66a37c 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1328,10 +1328,27 @@ struct sctp_inithdr_host { __u32 initial_tsn; }; +struct sctp_stream_priorities { + /* List of priorities scheduled */ + struct list_head prio_sched; + /* List of streams scheduled */ + struct list_head active; + /* The next stream stream in line */ + struct sctp_stream_out_ext *next; + __u16 prio; +}; + struct sctp_stream_out_ext { __u64 abandoned_unsent[SCTP_PR_INDEX(MAX) + 1]; __u64 abandoned_sent[SCTP_PR_INDEX(MAX) + 1]; struct list_head outq; /* chunks enqueued by this stream */ + union { + struct { + /* Scheduled streams list */ + struct list_head prio_list; + struct sctp_stream_priorities *prio_head; + }; + }; }; struct sctp_stream_out { @@ -1351,6 +1368,13 @@ struct sctp_stream { __u16 incnt; /* Current stream being sent, if any */ struct sctp_stream_out *out_curr; + union { + /* Fields used by priority scheduler */ + struct { + /* List of priorities scheduled */ + struct list_head prio_list; + }; + }; }; #define SCTP_STREAM_CLOSED 0x00 diff --git a/include/uapi/linux/sctp.h b/include/uapi/linux/sctp.h index 00ac417d2c4f..850fa8b29d7e 100644 --- a/include/uapi/linux/sctp.h +++ b/include/uapi/linux/sctp.h @@ -1099,7 +1099,8 @@ struct sctp_add_streams { /* SCTP Stream schedulers */ enum sctp_sched_type { SCTP_SS_FCFS, - SCTP_SS_MAX = SCTP_SS_FCFS + SCTP_SS_PRIO, + SCTP_SS_MAX = SCTP_SS_PRIO }; #endif /* _UAPI_SCTP_H */ diff --git a/net/sctp/Makefile b/net/sctp/Makefile index 0f6e6d1d69fd..647c9cfd4e95 100644 --- a/net/sctp/Makefile +++ b/net/sctp/Makefile @@ -12,7 +12,7 @@ sctp-y := sm_statetable.o sm_statefuns.o sm_sideeffect.o \ inqueue.o outqueue.o ulpqueue.o \ tsnmap.o bind_addr.o socket.o primitive.o \ output.o input.o debug.o stream.o auth.o \ - offload.o stream_sched.o + offload.o stream_sched.o stream_sched_prio.o sctp_probe-y := probe.o diff --git a/net/sctp/stream_sched.c b/net/sctp/stream_sched.c index 40a9a9de2b98..115ddb765169 100644 --- a/net/sctp/stream_sched.c +++ b/net/sctp/stream_sched.c @@ -121,8 +121,11 @@ static struct sctp_sched_ops sctp_sched_fcfs = { /* API to other parts of the stack */ +extern struct sctp_sched_ops sctp_sched_prio; + struct sctp_sched_ops *sctp_sched_ops[] = { &sctp_sched_fcfs, + &sctp_sched_prio, }; int sctp_sched_set_sched(struct sctp_association *asoc, diff --git a/net/sctp/stream_sched_prio.c b/net/sctp/stream_sched_prio.c new file mode 100644 index 000000000000..384dbf3c8760 --- /dev/null +++ b/net/sctp/stream_sched_prio.c @@ -0,0 +1,347 @@ +/* SCTP kernel implementation + * (C) Copyright Red Hat Inc. 2017 + * + * This file is part of the SCTP kernel implementation + * + * These functions manipulate sctp stream queue/scheduling. + * + * This SCTP implementation is free software; + * you can redistribute it and/or modify it under the terms of + * the GNU General Public License as published by + * the Free Software Foundation; either version 2, or (at your option) + * any later version. + * + * This SCTP implementation 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 GNU CC; see the file COPYING. If not, see + * . + * + * Please send any bug reports or fixes you make to the + * email addresched(es): + * lksctp developers + * + * Written or modified by: + * Marcelo Ricardo Leitner + */ + +#include +#include +#include +#include + +/* Priority handling + * RFC DRAFT ndata section 3.4 + */ + +static void sctp_sched_prio_unsched_all(struct sctp_stream *stream); + +static struct sctp_stream_priorities *sctp_sched_prio_new_head( + struct sctp_stream *stream, int prio, gfp_t gfp) +{ + struct sctp_stream_priorities *p; + + p = kmalloc(sizeof(*p), gfp); + if (!p) + return NULL; + + INIT_LIST_HEAD(&p->prio_sched); + INIT_LIST_HEAD(&p->active); + p->next = NULL; + p->prio = prio; + + return p; +} + +static struct sctp_stream_priorities *sctp_sched_prio_get_head( + struct sctp_stream *stream, int prio, gfp_t gfp) +{ + struct sctp_stream_priorities *p; + int i; + + /* Look into scheduled priorities first, as they are sorted and + * we can find it fast IF it's scheduled. + */ + list_for_each_entry(p, &stream->prio_list, prio_sched) { + if (p->prio == prio) + return p; + if (p->prio > prio) + break; + } + + /* No luck. So we search on all streams now. */ + for (i = 0; i < stream->outcnt; i++) { + if (!stream->out[i].ext) + continue; + + p = stream->out[i].ext->prio_head; + if (!p) + /* Means all other streams won't be initialized + * as well. + */ + break; + if (p->prio == prio) + return p; + } + + /* If not even there, allocate a new one. */ + return sctp_sched_prio_new_head(stream, prio, gfp); +} + +static void sctp_sched_prio_next_stream(struct sctp_stream_priorities *p) +{ + struct list_head *pos; + + pos = p->next->prio_list.next; + if (pos == &p->active) + pos = pos->next; + p->next = list_entry(pos, struct sctp_stream_out_ext, prio_list); +} + +static bool sctp_sched_prio_unsched(struct sctp_stream_out_ext *soute) +{ + bool scheduled = false; + + if (!list_empty(&soute->prio_list)) { + struct sctp_stream_priorities *prio_head = soute->prio_head; + + /* Scheduled */ + scheduled = true; + + if (prio_head->next == soute) + /* Try to move to the next stream */ + sctp_sched_prio_next_stream(prio_head); + + list_del_init(&soute->prio_list); + + /* Also unsched the priority if this was the last stream */ + if (list_empty(&prio_head->active)) { + list_del_init(&prio_head->prio_sched); + /* If there is no stream left, clear next */ + prio_head->next = NULL; + } + } + + return scheduled; +} + +static void sctp_sched_prio_sched(struct sctp_stream *stream, + struct sctp_stream_out_ext *soute) +{ + struct sctp_stream_priorities *prio, *prio_head; + + prio_head = soute->prio_head; + + /* Nothing to do if already scheduled */ + if (!list_empty(&soute->prio_list)) + return; + + /* Schedule the stream. If there is a next, we schedule the new + * one before it, so it's the last in round robin order. + * If there isn't, we also have to schedule the priority. + */ + if (prio_head->next) { + list_add(&soute->prio_list, prio_head->next->prio_list.prev); + return; + } + + list_add(&soute->prio_list, &prio_head->active); + prio_head->next = soute; + + list_for_each_entry(prio, &stream->prio_list, prio_sched) { + if (prio->prio > prio_head->prio) { + list_add(&prio_head->prio_sched, prio->prio_sched.prev); + return; + } + } + + list_add_tail(&prio_head->prio_sched, &stream->prio_list); +} + +static int sctp_sched_prio_set(struct sctp_stream *stream, __u16 sid, + __u16 prio, gfp_t gfp) +{ + struct sctp_stream_out *sout = &stream->out[sid]; + struct sctp_stream_out_ext *soute = sout->ext; + struct sctp_stream_priorities *prio_head, *old; + bool reschedule = false; + int i; + + prio_head = sctp_sched_prio_get_head(stream, prio, gfp); + if (!prio_head) + return -ENOMEM; + + reschedule = sctp_sched_prio_unsched(soute); + old = soute->prio_head; + soute->prio_head = prio_head; + if (reschedule) + sctp_sched_prio_sched(stream, soute); + + if (!old) + /* Happens when we set the priority for the first time */ + return 0; + + for (i = 0; i < stream->outcnt; i++) { + soute = stream->out[i].ext; + if (soute && soute->prio_head == old) + /* It's still in use, nothing else to do here. */ + return 0; + } + + /* No hits, we are good to free it. */ + kfree(old); + + return 0; +} + +static int sctp_sched_prio_get(struct sctp_stream *stream, __u16 sid, + __u16 *value) +{ + *value = stream->out[sid].ext->prio_head->prio; + return 0; +} + +static int sctp_sched_prio_init(struct sctp_stream *stream) +{ + INIT_LIST_HEAD(&stream->prio_list); + + return 0; +} + +static int sctp_sched_prio_init_sid(struct sctp_stream *stream, __u16 sid, + gfp_t gfp) +{ + INIT_LIST_HEAD(&stream->out[sid].ext->prio_list); + return sctp_sched_prio_set(stream, sid, 0, gfp); +} + +static void sctp_sched_prio_free(struct sctp_stream *stream) +{ + struct sctp_stream_priorities *prio, *n; + LIST_HEAD(list); + int i; + + /* As we don't keep a list of priorities, to avoid multiple + * frees we have to do it in 3 steps: + * 1. unsched everyone, so the lists are free to use in 2. + * 2. build the list of the priorities + * 3. free the list + */ + sctp_sched_prio_unsched_all(stream); + for (i = 0; i < stream->outcnt; i++) { + if (!stream->out[i].ext) + continue; + prio = stream->out[i].ext->prio_head; + if (prio && list_empty(&prio->prio_sched)) + list_add(&prio->prio_sched, &list); + } + list_for_each_entry_safe(prio, n, &list, prio_sched) { + list_del_init(&prio->prio_sched); + kfree(prio); + } +} + +static void sctp_sched_prio_enqueue(struct sctp_outq *q, + struct sctp_datamsg *msg) +{ + struct sctp_stream *stream; + struct sctp_chunk *ch; + __u16 sid; + + ch = list_first_entry(&msg->chunks, struct sctp_chunk, frag_list); + sid = sctp_chunk_stream_no(ch); + stream = &q->asoc->stream; + sctp_sched_prio_sched(stream, stream->out[sid].ext); +} + +static struct sctp_chunk *sctp_sched_prio_dequeue(struct sctp_outq *q) +{ + struct sctp_stream *stream = &q->asoc->stream; + struct sctp_stream_priorities *prio; + struct sctp_stream_out_ext *soute; + struct sctp_chunk *ch = NULL; + + /* Bail out quickly if queue is empty */ + if (list_empty(&q->out_chunk_list)) + goto out; + + /* Find which chunk is next. It's easy, it's either the current + * one or the first chunk on the next active stream. + */ + if (stream->out_curr) { + soute = stream->out_curr->ext; + } else { + prio = list_entry(stream->prio_list.next, + struct sctp_stream_priorities, prio_sched); + soute = prio->next; + } + ch = list_entry(soute->outq.next, struct sctp_chunk, stream_list); + sctp_sched_dequeue_common(q, ch); + +out: + return ch; +} + +static void sctp_sched_prio_dequeue_done(struct sctp_outq *q, + struct sctp_chunk *ch) +{ + struct sctp_stream_priorities *prio; + struct sctp_stream_out_ext *soute; + __u16 sid; + + /* Last chunk on that msg, move to the next stream on + * this priority. + */ + sid = sctp_chunk_stream_no(ch); + soute = q->asoc->stream.out[sid].ext; + prio = soute->prio_head; + + sctp_sched_prio_next_stream(prio); + + if (list_empty(&soute->outq)) + sctp_sched_prio_unsched(soute); +} + +static void sctp_sched_prio_sched_all(struct sctp_stream *stream) +{ + struct sctp_association *asoc; + struct sctp_stream_out *sout; + struct sctp_chunk *ch; + + asoc = container_of(stream, struct sctp_association, stream); + list_for_each_entry(ch, &asoc->outqueue.out_chunk_list, list) { + __u16 sid; + + sid = sctp_chunk_stream_no(ch); + sout = &stream->out[sid]; + if (sout->ext) + sctp_sched_prio_sched(stream, sout->ext); + } +} + +static void sctp_sched_prio_unsched_all(struct sctp_stream *stream) +{ + struct sctp_stream_priorities *p, *tmp; + struct sctp_stream_out_ext *soute, *souttmp; + + list_for_each_entry_safe(p, tmp, &stream->prio_list, prio_sched) + list_for_each_entry_safe(soute, souttmp, &p->active, prio_list) + sctp_sched_prio_unsched(soute); +} + +struct sctp_sched_ops sctp_sched_prio = { + .set = sctp_sched_prio_set, + .get = sctp_sched_prio_get, + .init = sctp_sched_prio_init, + .init_sid = sctp_sched_prio_init_sid, + .free = sctp_sched_prio_free, + .enqueue = sctp_sched_prio_enqueue, + .dequeue = sctp_sched_prio_dequeue, + .dequeue_done = sctp_sched_prio_dequeue_done, + .sched_all = sctp_sched_prio_sched_all, + .unsched_all = sctp_sched_prio_unsched_all, +}; -- cgit v1.2.3 From ac1ed8b82cd60ba8e7d84103ac1414b8c577c485 Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Tue, 3 Oct 2017 19:20:17 -0300 Subject: sctp: introduce round robin stream scheduler This patch introduces RFC Draft ndata section 3.2 Priority Based Scheduler (SCTP_SS_RR). Works by maintaining a list of enqueued streams and tracking the last one used to send data. When the datamsg is done, it switches to the next stream. See-also: https://tools.ietf.org/html/draft-ietf-tsvwg-sctp-ndata-13 Tested-by: Xin Long Signed-off-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 11 +++ include/uapi/linux/sctp.h | 3 +- net/sctp/Makefile | 3 +- net/sctp/stream_sched.c | 2 + net/sctp/stream_sched_rr.c | 201 +++++++++++++++++++++++++++++++++++++++++++++ 5 files changed, 218 insertions(+), 2 deletions(-) create mode 100644 net/sctp/stream_sched_rr.c (limited to 'include/net') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 40eb8d66a37c..16f949eef52f 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1348,6 +1348,10 @@ struct sctp_stream_out_ext { struct list_head prio_list; struct sctp_stream_priorities *prio_head; }; + /* Fields used by RR scheduler */ + struct { + struct list_head rr_list; + }; }; }; @@ -1374,6 +1378,13 @@ struct sctp_stream { /* List of priorities scheduled */ struct list_head prio_list; }; + /* Fields used by RR scheduler */ + struct { + /* List of streams scheduled */ + struct list_head rr_list; + /* The next stream stream in line */ + struct sctp_stream_out_ext *rr_next; + }; }; }; diff --git a/include/uapi/linux/sctp.h b/include/uapi/linux/sctp.h index 850fa8b29d7e..6cd7d416ca40 100644 --- a/include/uapi/linux/sctp.h +++ b/include/uapi/linux/sctp.h @@ -1100,7 +1100,8 @@ struct sctp_add_streams { enum sctp_sched_type { SCTP_SS_FCFS, SCTP_SS_PRIO, - SCTP_SS_MAX = SCTP_SS_PRIO + SCTP_SS_RR, + SCTP_SS_MAX = SCTP_SS_RR }; #endif /* _UAPI_SCTP_H */ diff --git a/net/sctp/Makefile b/net/sctp/Makefile index 647c9cfd4e95..bf90c5397719 100644 --- a/net/sctp/Makefile +++ b/net/sctp/Makefile @@ -12,7 +12,8 @@ sctp-y := sm_statetable.o sm_statefuns.o sm_sideeffect.o \ inqueue.o outqueue.o ulpqueue.o \ tsnmap.o bind_addr.o socket.o primitive.o \ output.o input.o debug.o stream.o auth.o \ - offload.o stream_sched.o stream_sched_prio.o + offload.o stream_sched.o stream_sched_prio.o \ + stream_sched_rr.o sctp_probe-y := probe.o diff --git a/net/sctp/stream_sched.c b/net/sctp/stream_sched.c index 115ddb765169..03513a9fa110 100644 --- a/net/sctp/stream_sched.c +++ b/net/sctp/stream_sched.c @@ -122,10 +122,12 @@ static struct sctp_sched_ops sctp_sched_fcfs = { /* API to other parts of the stack */ extern struct sctp_sched_ops sctp_sched_prio; +extern struct sctp_sched_ops sctp_sched_rr; struct sctp_sched_ops *sctp_sched_ops[] = { &sctp_sched_fcfs, &sctp_sched_prio, + &sctp_sched_rr, }; int sctp_sched_set_sched(struct sctp_association *asoc, diff --git a/net/sctp/stream_sched_rr.c b/net/sctp/stream_sched_rr.c new file mode 100644 index 000000000000..7612a438c5b9 --- /dev/null +++ b/net/sctp/stream_sched_rr.c @@ -0,0 +1,201 @@ +/* SCTP kernel implementation + * (C) Copyright Red Hat Inc. 2017 + * + * This file is part of the SCTP kernel implementation + * + * These functions manipulate sctp stream queue/scheduling. + * + * This SCTP implementation is free software; + * you can redistribute it and/or modify it under the terms of + * the GNU General Public License as published by + * the Free Software Foundation; either version 2, or (at your option) + * any later version. + * + * This SCTP implementation 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 GNU CC; see the file COPYING. If not, see + * . + * + * Please send any bug reports or fixes you make to the + * email addresched(es): + * lksctp developers + * + * Written or modified by: + * Marcelo Ricardo Leitner + */ + +#include +#include +#include +#include + +/* Priority handling + * RFC DRAFT ndata section 3.2 + */ +static void sctp_sched_rr_unsched_all(struct sctp_stream *stream); + +static void sctp_sched_rr_next_stream(struct sctp_stream *stream) +{ + struct list_head *pos; + + pos = stream->rr_next->rr_list.next; + if (pos == &stream->rr_list) + pos = pos->next; + stream->rr_next = list_entry(pos, struct sctp_stream_out_ext, rr_list); +} + +static void sctp_sched_rr_unsched(struct sctp_stream *stream, + struct sctp_stream_out_ext *soute) +{ + if (stream->rr_next == soute) + /* Try to move to the next stream */ + sctp_sched_rr_next_stream(stream); + + list_del_init(&soute->rr_list); + + /* If we have no other stream queued, clear next */ + if (list_empty(&stream->rr_list)) + stream->rr_next = NULL; +} + +static void sctp_sched_rr_sched(struct sctp_stream *stream, + struct sctp_stream_out_ext *soute) +{ + if (!list_empty(&soute->rr_list)) + /* Already scheduled. */ + return; + + /* Schedule the stream */ + list_add_tail(&soute->rr_list, &stream->rr_list); + + if (!stream->rr_next) + stream->rr_next = soute; +} + +static int sctp_sched_rr_set(struct sctp_stream *stream, __u16 sid, + __u16 prio, gfp_t gfp) +{ + return 0; +} + +static int sctp_sched_rr_get(struct sctp_stream *stream, __u16 sid, + __u16 *value) +{ + return 0; +} + +static int sctp_sched_rr_init(struct sctp_stream *stream) +{ + INIT_LIST_HEAD(&stream->rr_list); + stream->rr_next = NULL; + + return 0; +} + +static int sctp_sched_rr_init_sid(struct sctp_stream *stream, __u16 sid, + gfp_t gfp) +{ + INIT_LIST_HEAD(&stream->out[sid].ext->rr_list); + + return 0; +} + +static void sctp_sched_rr_free(struct sctp_stream *stream) +{ + sctp_sched_rr_unsched_all(stream); +} + +static void sctp_sched_rr_enqueue(struct sctp_outq *q, + struct sctp_datamsg *msg) +{ + struct sctp_stream *stream; + struct sctp_chunk *ch; + __u16 sid; + + ch = list_first_entry(&msg->chunks, struct sctp_chunk, frag_list); + sid = sctp_chunk_stream_no(ch); + stream = &q->asoc->stream; + sctp_sched_rr_sched(stream, stream->out[sid].ext); +} + +static struct sctp_chunk *sctp_sched_rr_dequeue(struct sctp_outq *q) +{ + struct sctp_stream *stream = &q->asoc->stream; + struct sctp_stream_out_ext *soute; + struct sctp_chunk *ch = NULL; + + /* Bail out quickly if queue is empty */ + if (list_empty(&q->out_chunk_list)) + goto out; + + /* Find which chunk is next */ + if (stream->out_curr) + soute = stream->out_curr->ext; + else + soute = stream->rr_next; + ch = list_entry(soute->outq.next, struct sctp_chunk, stream_list); + + sctp_sched_dequeue_common(q, ch); + +out: + return ch; +} + +static void sctp_sched_rr_dequeue_done(struct sctp_outq *q, + struct sctp_chunk *ch) +{ + struct sctp_stream_out_ext *soute; + __u16 sid; + + /* Last chunk on that msg, move to the next stream */ + sid = sctp_chunk_stream_no(ch); + soute = q->asoc->stream.out[sid].ext; + + sctp_sched_rr_next_stream(&q->asoc->stream); + + if (list_empty(&soute->outq)) + sctp_sched_rr_unsched(&q->asoc->stream, soute); +} + +static void sctp_sched_rr_sched_all(struct sctp_stream *stream) +{ + struct sctp_association *asoc; + struct sctp_stream_out_ext *soute; + struct sctp_chunk *ch; + + asoc = container_of(stream, struct sctp_association, stream); + list_for_each_entry(ch, &asoc->outqueue.out_chunk_list, list) { + __u16 sid; + + sid = sctp_chunk_stream_no(ch); + soute = stream->out[sid].ext; + if (soute) + sctp_sched_rr_sched(stream, soute); + } +} + +static void sctp_sched_rr_unsched_all(struct sctp_stream *stream) +{ + struct sctp_stream_out_ext *soute, *tmp; + + list_for_each_entry_safe(soute, tmp, &stream->rr_list, rr_list) + sctp_sched_rr_unsched(stream, soute); +} + +struct sctp_sched_ops sctp_sched_rr = { + .set = sctp_sched_rr_set, + .get = sctp_sched_rr_get, + .init = sctp_sched_rr_init, + .init_sid = sctp_sched_rr_init_sid, + .free = sctp_sched_rr_free, + .enqueue = sctp_sched_rr_enqueue, + .dequeue = sctp_sched_rr_dequeue, + .dequeue_done = sctp_sched_rr_dequeue_done, + .sched_all = sctp_sched_rr_sched_all, + .unsched_all = sctp_sched_rr_unsched_all, +}; -- cgit v1.2.3 From e774d96b7d2c3489bfb5bbdc2b65ed41cd68d3d5 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 4 Oct 2017 15:55:29 +0200 Subject: rtnetlink: remove slave_validate callback no users in the tree. Signed-off-by: Florian Westphal Acked-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/rtnetlink.h | 3 --- net/core/rtnetlink.c | 6 ------ 2 files changed, 9 deletions(-) (limited to 'include/net') diff --git a/include/net/rtnetlink.h b/include/net/rtnetlink.h index 21837ca68ecc..6520993ff449 100644 --- a/include/net/rtnetlink.h +++ b/include/net/rtnetlink.h @@ -93,9 +93,6 @@ struct rtnl_link_ops { int slave_maxtype; const struct nla_policy *slave_policy; - int (*slave_validate)(struct nlattr *tb[], - struct nlattr *data[], - struct netlink_ext_ack *extack); int (*slave_changelink)(struct net_device *dev, struct net_device *slave_dev, struct nlattr *tb[], diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 3961f87cdc76..b63c5759641f 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -2631,12 +2631,6 @@ replay: return err; slave_data = slave_attr; } - if (m_ops->slave_validate) { - err = m_ops->slave_validate(tb, slave_data, - extack); - if (err < 0) - return err; - } } if (dev) { -- cgit v1.2.3 From 5c45121dc39026ab2139910e57cf933fd57d30f2 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 4 Oct 2017 15:58:49 +0200 Subject: rtnetlink: remove __rtnl_af_unregister switch the only caller to rtnl_af_unregister. Signed-off-by: Florian Westphal Signed-off-by: David S. Miller --- include/net/rtnetlink.h | 2 -- net/core/rtnetlink.c | 14 +------------- net/ipv6/addrconf.c | 4 ++-- 3 files changed, 3 insertions(+), 17 deletions(-) (limited to 'include/net') diff --git a/include/net/rtnetlink.h b/include/net/rtnetlink.h index 6520993ff449..e3ca8e2e3103 100644 --- a/include/net/rtnetlink.h +++ b/include/net/rtnetlink.h @@ -151,8 +151,6 @@ struct rtnl_af_ops { size_t (*get_stats_af_size)(const struct net_device *dev); }; -void __rtnl_af_unregister(struct rtnl_af_ops *ops); - void rtnl_af_register(struct rtnl_af_ops *ops); void rtnl_af_unregister(struct rtnl_af_ops *ops); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index b63c5759641f..3fb1ca33cba4 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -475,18 +475,6 @@ void rtnl_af_register(struct rtnl_af_ops *ops) } EXPORT_SYMBOL_GPL(rtnl_af_register); -/** - * __rtnl_af_unregister - Unregister rtnl_af_ops from rtnetlink. - * @ops: struct rtnl_af_ops * to unregister - * - * The caller must hold the rtnl_mutex. - */ -void __rtnl_af_unregister(struct rtnl_af_ops *ops) -{ - list_del(&ops->list); -} -EXPORT_SYMBOL_GPL(__rtnl_af_unregister); - /** * rtnl_af_unregister - Unregister rtnl_af_ops from rtnetlink. * @ops: struct rtnl_af_ops * to unregister @@ -494,7 +482,7 @@ EXPORT_SYMBOL_GPL(__rtnl_af_unregister); void rtnl_af_unregister(struct rtnl_af_ops *ops) { rtnl_lock(); - __rtnl_af_unregister(ops); + list_del(&ops->list); rtnl_unlock(); } EXPORT_SYMBOL_GPL(rtnl_af_unregister); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index f553f72d0bee..837418ff2d4b 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -6618,9 +6618,9 @@ void addrconf_cleanup(void) unregister_pernet_subsys(&addrconf_ops); ipv6_addr_label_cleanup(); - rtnl_lock(); + rtnl_af_unregister(&inet6_ops); - __rtnl_af_unregister(&inet6_ops); + rtnl_lock(); /* clean dev list */ for_each_netdev(&init_net, dev) { -- cgit v1.2.3 From 33eaf2a6eb48ebf00374aaaf4b1b43f9950dcbe4 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Wed, 4 Oct 2017 17:48:46 -0700 Subject: net: Add extack to ndo_add_slave Pass extack to do_set_master and down to ndo_add_slave Signed-off-by: David Ahern Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 5 +++-- drivers/net/bonding/bond_options.c | 2 +- drivers/net/team/team.c | 3 ++- drivers/net/vrf.c | 3 ++- include/linux/netdevice.h | 3 ++- include/net/bonding.h | 3 ++- net/batman-adv/soft-interface.c | 3 ++- net/bridge/br_device.c | 3 ++- net/core/rtnetlink.c | 10 ++++++---- 9 files changed, 22 insertions(+), 13 deletions(-) (limited to 'include/net') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index b19dc033fb36..78feb94a36db 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1328,7 +1328,8 @@ void bond_lower_state_changed(struct slave *slave) } /* enslave device to bond device */ -int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev) +int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev, + struct netlink_ext_ack *extack) { struct bonding *bond = netdev_priv(bond_dev); const struct net_device_ops *slave_ops = slave_dev->netdev_ops; @@ -3492,7 +3493,7 @@ static int bond_do_ioctl(struct net_device *bond_dev, struct ifreq *ifr, int cmd switch (cmd) { case BOND_ENSLAVE_OLD: case SIOCBONDENSLAVE: - res = bond_enslave(bond_dev, slave_dev); + res = bond_enslave(bond_dev, slave_dev, NULL); break; case BOND_RELEASE_OLD: case SIOCBONDRELEASE: diff --git a/drivers/net/bonding/bond_options.c b/drivers/net/bonding/bond_options.c index 5931aa2fe997..8a9b085c2a98 100644 --- a/drivers/net/bonding/bond_options.c +++ b/drivers/net/bonding/bond_options.c @@ -1383,7 +1383,7 @@ static int bond_option_slaves_set(struct bonding *bond, switch (command[0]) { case '+': netdev_dbg(bond->dev, "Adding slave %s\n", dev->name); - ret = bond_enslave(bond->dev, dev); + ret = bond_enslave(bond->dev, dev, NULL); break; case '-': diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index ae53e899259f..4359d45aa131 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -1914,7 +1914,8 @@ static int team_netpoll_setup(struct net_device *dev, } #endif -static int team_add_slave(struct net_device *dev, struct net_device *port_dev) +static int team_add_slave(struct net_device *dev, struct net_device *port_dev, + struct netlink_ext_ack *extack) { struct team *team = netdev_priv(dev); int err; diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index cc18b7b11612..4a082ef53533 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -788,7 +788,8 @@ err: return ret; } -static int vrf_add_slave(struct net_device *dev, struct net_device *port_dev) +static int vrf_add_slave(struct net_device *dev, struct net_device *port_dev, + struct netlink_ext_ack *extack) { if (netif_is_l3_master(port_dev) || netif_is_l3_slave(port_dev)) return -EINVAL; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 05fcaba4b0d9..368a5064a487 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1246,7 +1246,8 @@ struct net_device_ops { u32 flow_id); #endif int (*ndo_add_slave)(struct net_device *dev, - struct net_device *slave_dev); + struct net_device *slave_dev, + struct netlink_ext_ack *extack); int (*ndo_del_slave)(struct net_device *dev, struct net_device *slave_dev); netdev_features_t (*ndo_fix_features)(struct net_device *dev, diff --git a/include/net/bonding.h b/include/net/bonding.h index b2e68657a216..2860cc66c2bb 100644 --- a/include/net/bonding.h +++ b/include/net/bonding.h @@ -596,7 +596,8 @@ void bond_destroy_sysfs(struct bond_net *net); void bond_prepare_sysfs_group(struct bonding *bond); int bond_sysfs_slave_add(struct slave *slave); void bond_sysfs_slave_del(struct slave *slave); -int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev); +int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev, + struct netlink_ext_ack *extack); int bond_release(struct net_device *bond_dev, struct net_device *slave_dev); u32 bond_xmit_hash(struct bonding *bond, struct sk_buff *skb); int bond_set_carrier(struct bonding *bond); diff --git a/net/batman-adv/soft-interface.c b/net/batman-adv/soft-interface.c index c2c986746d0b..e7d5fbb6ad53 100644 --- a/net/batman-adv/soft-interface.c +++ b/net/batman-adv/soft-interface.c @@ -867,7 +867,8 @@ free_bat_counters: * Return: 0 if successful or error otherwise. */ static int batadv_softif_slave_add(struct net_device *dev, - struct net_device *slave_dev) + struct net_device *slave_dev, + struct netlink_ext_ack *extack) { struct batadv_hard_iface *hard_iface; struct net *net = dev_net(dev); diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index f6b6a92f1c48..cb0131d70ab1 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -320,7 +320,8 @@ void br_netpoll_disable(struct net_bridge_port *p) #endif -static int br_add_slave(struct net_device *dev, struct net_device *slave_dev) +static int br_add_slave(struct net_device *dev, struct net_device *slave_dev, + struct netlink_ext_ack *extack) { struct net_bridge *br = netdev_priv(dev); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 1ee98b1369d5..c5ee429bcce9 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1957,7 +1957,8 @@ static int do_setvfinfo(struct net_device *dev, struct nlattr **tb) return err; } -static int do_set_master(struct net_device *dev, int ifindex) +static int do_set_master(struct net_device *dev, int ifindex, + struct netlink_ext_ack *extack) { struct net_device *upper_dev = netdev_master_upper_dev_get(dev); const struct net_device_ops *ops; @@ -1982,7 +1983,7 @@ static int do_set_master(struct net_device *dev, int ifindex) return -EINVAL; ops = upper_dev->netdev_ops; if (ops->ndo_add_slave) { - err = ops->ndo_add_slave(upper_dev, dev); + err = ops->ndo_add_slave(upper_dev, dev, extack); if (err) return err; } else { @@ -2115,7 +2116,7 @@ static int do_setlink(const struct sk_buff *skb, } if (tb[IFLA_MASTER]) { - err = do_set_master(dev, nla_get_u32(tb[IFLA_MASTER])); + err = do_set_master(dev, nla_get_u32(tb[IFLA_MASTER]), extack); if (err) goto errout; status |= DO_SETLINK_MODIFIED; @@ -2753,7 +2754,8 @@ replay: goto out_unregister; } if (tb[IFLA_MASTER]) { - err = do_set_master(dev, nla_get_u32(tb[IFLA_MASTER])); + err = do_set_master(dev, nla_get_u32(tb[IFLA_MASTER]), + extack); if (err) goto out_unregister; } -- cgit v1.2.3 From 44f209807ee87a5eddf6c0432f3fb63cec27bad8 Mon Sep 17 00:00:00 2001 From: Stefan Hajnoczi Date: Thu, 5 Oct 2017 16:46:50 -0400 Subject: VSOCK: export socket tables for sock_diag interface The socket table symbols need to be exported from vsock.ko so that the vsock_diag.ko module will be able to traverse sockets. Signed-off-by: Stefan Hajnoczi Signed-off-by: David S. Miller --- include/net/af_vsock.h | 5 +++++ net/vmw_vsock/af_vsock.c | 10 ++++++---- 2 files changed, 11 insertions(+), 4 deletions(-) (limited to 'include/net') diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h index f9fb566e75cf..30cba806e344 100644 --- a/include/net/af_vsock.h +++ b/include/net/af_vsock.h @@ -27,6 +27,11 @@ #define LAST_RESERVED_PORT 1023 +#define VSOCK_HASH_SIZE 251 +extern struct list_head vsock_bind_table[VSOCK_HASH_SIZE + 1]; +extern struct list_head vsock_connected_table[VSOCK_HASH_SIZE]; +extern spinlock_t vsock_table_lock; + #define vsock_sk(__sk) ((struct vsock_sock *)__sk) #define sk_vsock(__vsk) (&(__vsk)->sk) diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index dfc8c51e4d74..9afe4da8c67d 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -153,7 +153,6 @@ EXPORT_SYMBOL_GPL(vm_sockets_get_local_cid); * vsock_bind_table[VSOCK_HASH_SIZE] is for unbound sockets. The hash function * mods with VSOCK_HASH_SIZE to ensure this. */ -#define VSOCK_HASH_SIZE 251 #define MAX_PORT_RETRIES 24 #define VSOCK_HASH(addr) ((addr)->svm_port % VSOCK_HASH_SIZE) @@ -168,9 +167,12 @@ EXPORT_SYMBOL_GPL(vm_sockets_get_local_cid); #define vsock_connected_sockets_vsk(vsk) \ vsock_connected_sockets(&(vsk)->remote_addr, &(vsk)->local_addr) -static struct list_head vsock_bind_table[VSOCK_HASH_SIZE + 1]; -static struct list_head vsock_connected_table[VSOCK_HASH_SIZE]; -static DEFINE_SPINLOCK(vsock_table_lock); +struct list_head vsock_bind_table[VSOCK_HASH_SIZE + 1]; +EXPORT_SYMBOL_GPL(vsock_bind_table); +struct list_head vsock_connected_table[VSOCK_HASH_SIZE]; +EXPORT_SYMBOL_GPL(vsock_connected_table); +DEFINE_SPINLOCK(vsock_table_lock); +EXPORT_SYMBOL_GPL(vsock_table_lock); /* Autobind this socket to the local address if necessary. */ static int vsock_auto_bind(struct vsock_sock *vsk) -- cgit v1.2.3 From bf359b8127719535f88494adb3c2b73c06667dcd Mon Sep 17 00:00:00 2001 From: Stefan Hajnoczi Date: Thu, 5 Oct 2017 16:46:51 -0400 Subject: VSOCK: move __vsock_in_bound/connected_table() to af_vsock.h The vsock_diag.ko module will need to check socket table membership. Signed-off-by: Stefan Hajnoczi Signed-off-by: David S. Miller --- include/net/af_vsock.h | 12 ++++++++++++ net/vmw_vsock/af_vsock.c | 10 ---------- 2 files changed, 12 insertions(+), 10 deletions(-) (limited to 'include/net') diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h index 30cba806e344..3dd217718a2f 100644 --- a/include/net/af_vsock.h +++ b/include/net/af_vsock.h @@ -180,6 +180,18 @@ const struct vsock_transport *vsock_core_get_transport(void); /**** UTILS ****/ +/* vsock_table_lock must be held */ +static inline bool __vsock_in_bound_table(struct vsock_sock *vsk) +{ + return !list_empty(&vsk->bound_table); +} + +/* vsock_table_lock must be held */ +static inline bool __vsock_in_connected_table(struct vsock_sock *vsk) +{ + return !list_empty(&vsk->connected_table); +} + void vsock_release_pending(struct sock *pending); void vsock_add_pending(struct sock *listener, struct sock *pending); void vsock_remove_pending(struct sock *listener, struct sock *pending); diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 9afe4da8c67d..9b179a0081b3 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -250,16 +250,6 @@ static struct sock *__vsock_find_connected_socket(struct sockaddr_vm *src, return NULL; } -static bool __vsock_in_bound_table(struct vsock_sock *vsk) -{ - return !list_empty(&vsk->bound_table); -} - -static bool __vsock_in_connected_table(struct vsock_sock *vsk) -{ - return !list_empty(&vsk->connected_table); -} - static void vsock_insert_unbound(struct vsock_sock *vsk) { spin_lock_bh(&vsock_table_lock); -- cgit v1.2.3 From 3b4477d2dcf2709d0be89e2a8dced3d0f4a017f2 Mon Sep 17 00:00:00 2001 From: Stefan Hajnoczi Date: Thu, 5 Oct 2017 16:46:52 -0400 Subject: VSOCK: use TCP state constants for sk_state There are two state fields: socket->state and sock->sk_state. The socket->state field uses SS_UNCONNECTED, SS_CONNECTED, etc while the sock->sk_state typically uses values that match TCP state constants (TCP_CLOSE, TCP_ESTABLISHED). AF_VSOCK does not follow this convention and instead uses SS_* constants for both fields. The sk_state field will be exposed to userspace through the vsock_diag interface for ss(8), netstat(8), and other programs. This patch switches sk_state to TCP state constants so that the meaning of this field is consistent with other address families. Not just AF_INET and AF_INET6 use the TCP constants, AF_UNIX and others do too. The following mapping was used to convert the code: SS_FREE -> TCP_CLOSE SS_UNCONNECTED -> TCP_CLOSE SS_CONNECTING -> TCP_SYN_SENT SS_CONNECTED -> TCP_ESTABLISHED SS_DISCONNECTING -> TCP_CLOSING VSOCK_SS_LISTEN -> TCP_LISTEN In __vsock_create() the sk_state initialization was dropped because sock_init_data() already initializes sk_state to TCP_CLOSE. Signed-off-by: Stefan Hajnoczi Signed-off-by: David S. Miller --- include/net/af_vsock.h | 3 -- net/vmw_vsock/af_vsock.c | 46 ++++++++++++++++------------ net/vmw_vsock/hyperv_transport.c | 12 ++++---- net/vmw_vsock/virtio_transport.c | 2 +- net/vmw_vsock/virtio_transport_common.c | 22 ++++++------- net/vmw_vsock/vmci_transport.c | 34 ++++++++++---------- net/vmw_vsock/vmci_transport_notify.c | 2 +- net/vmw_vsock/vmci_transport_notify_qstate.c | 2 +- 8 files changed, 64 insertions(+), 59 deletions(-) (limited to 'include/net') diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h index 3dd217718a2f..9324ac2d9ff2 100644 --- a/include/net/af_vsock.h +++ b/include/net/af_vsock.h @@ -22,9 +22,6 @@ #include "vsock_addr.h" -/* vsock-specific sock->sk_state constants */ -#define VSOCK_SS_LISTEN 255 - #define LAST_RESERVED_PORT 1023 #define VSOCK_HASH_SIZE 251 diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 9b179a0081b3..98359c19522f 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -36,7 +36,7 @@ * not support simultaneous connects (two "client" sockets connecting). * * - "Server" sockets are referred to as listener sockets throughout this - * implementation because they are in the VSOCK_SS_LISTEN state. When a + * implementation because they are in the TCP_LISTEN state. When a * connection request is received (the second kind of socket mentioned above), * we create a new socket and refer to it as a pending socket. These pending * sockets are placed on the pending connection list of the listener socket. @@ -82,6 +82,15 @@ * argument, we must ensure the reference count is increased to ensure the * socket isn't freed before the function is run; the deferred function will * then drop the reference. + * + * - sk->sk_state uses the TCP state constants because they are widely used by + * other address families and exposed to userspace tools like ss(8): + * + * TCP_CLOSE - unconnected + * TCP_SYN_SENT - connecting + * TCP_ESTABLISHED - connected + * TCP_CLOSING - disconnecting + * TCP_LISTEN - listening */ #include @@ -477,7 +486,7 @@ void vsock_pending_work(struct work_struct *work) if (vsock_in_connected_table(vsk)) vsock_remove_connected(vsk); - sk->sk_state = SS_FREE; + sk->sk_state = TCP_CLOSE; out: release_sock(sk); @@ -617,7 +626,6 @@ struct sock *__vsock_create(struct net *net, sk->sk_destruct = vsock_sk_destruct; sk->sk_backlog_rcv = vsock_queue_rcv_skb; - sk->sk_state = 0; sock_reset_flag(sk, SOCK_DONE); INIT_LIST_HEAD(&vsk->bound_table); @@ -891,7 +899,7 @@ static unsigned int vsock_poll(struct file *file, struct socket *sock, /* Listening sockets that have connections in their accept * queue can be read. */ - if (sk->sk_state == VSOCK_SS_LISTEN + if (sk->sk_state == TCP_LISTEN && !vsock_is_accept_queue_empty(sk)) mask |= POLLIN | POLLRDNORM; @@ -920,7 +928,7 @@ static unsigned int vsock_poll(struct file *file, struct socket *sock, } /* Connected sockets that can produce data can be written. */ - if (sk->sk_state == SS_CONNECTED) { + if (sk->sk_state == TCP_ESTABLISHED) { if (!(sk->sk_shutdown & SEND_SHUTDOWN)) { bool space_avail_now = false; int ret = transport->notify_poll_out( @@ -942,7 +950,7 @@ static unsigned int vsock_poll(struct file *file, struct socket *sock, * POLLOUT|POLLWRNORM when peer is closed and nothing to read, * but local send is not shutdown. */ - if (sk->sk_state == SS_UNCONNECTED) { + if (sk->sk_state == TCP_CLOSE) { if (!(sk->sk_shutdown & SEND_SHUTDOWN)) mask |= POLLOUT | POLLWRNORM; @@ -1112,9 +1120,9 @@ static void vsock_connect_timeout(struct work_struct *work) sk = sk_vsock(vsk); lock_sock(sk); - if (sk->sk_state == SS_CONNECTING && + if (sk->sk_state == TCP_SYN_SENT && (sk->sk_shutdown != SHUTDOWN_MASK)) { - sk->sk_state = SS_UNCONNECTED; + sk->sk_state = TCP_CLOSE; sk->sk_err = ETIMEDOUT; sk->sk_error_report(sk); cancel = 1; @@ -1160,7 +1168,7 @@ static int vsock_stream_connect(struct socket *sock, struct sockaddr *addr, err = -EALREADY; break; default: - if ((sk->sk_state == VSOCK_SS_LISTEN) || + if ((sk->sk_state == TCP_LISTEN) || vsock_addr_cast(addr, addr_len, &remote_addr) != 0) { err = -EINVAL; goto out; @@ -1183,7 +1191,7 @@ static int vsock_stream_connect(struct socket *sock, struct sockaddr *addr, if (err) goto out; - sk->sk_state = SS_CONNECTING; + sk->sk_state = TCP_SYN_SENT; err = transport->connect(vsk); if (err < 0) @@ -1203,7 +1211,7 @@ static int vsock_stream_connect(struct socket *sock, struct sockaddr *addr, timeout = vsk->connect_timeout; prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); - while (sk->sk_state != SS_CONNECTED && sk->sk_err == 0) { + while (sk->sk_state != TCP_ESTABLISHED && sk->sk_err == 0) { if (flags & O_NONBLOCK) { /* If we're not going to block, we schedule a timeout * function to generate a timeout on the connection @@ -1226,13 +1234,13 @@ static int vsock_stream_connect(struct socket *sock, struct sockaddr *addr, if (signal_pending(current)) { err = sock_intr_errno(timeout); - sk->sk_state = SS_UNCONNECTED; + sk->sk_state = TCP_CLOSE; sock->state = SS_UNCONNECTED; vsock_transport_cancel_pkt(vsk); goto out_wait; } else if (timeout == 0) { err = -ETIMEDOUT; - sk->sk_state = SS_UNCONNECTED; + sk->sk_state = TCP_CLOSE; sock->state = SS_UNCONNECTED; vsock_transport_cancel_pkt(vsk); goto out_wait; @@ -1243,7 +1251,7 @@ static int vsock_stream_connect(struct socket *sock, struct sockaddr *addr, if (sk->sk_err) { err = -sk->sk_err; - sk->sk_state = SS_UNCONNECTED; + sk->sk_state = TCP_CLOSE; sock->state = SS_UNCONNECTED; } else { err = 0; @@ -1276,7 +1284,7 @@ static int vsock_accept(struct socket *sock, struct socket *newsock, int flags, goto out; } - if (listener->sk_state != VSOCK_SS_LISTEN) { + if (listener->sk_state != TCP_LISTEN) { err = -EINVAL; goto out; } @@ -1366,7 +1374,7 @@ static int vsock_listen(struct socket *sock, int backlog) } sk->sk_max_ack_backlog = backlog; - sk->sk_state = VSOCK_SS_LISTEN; + sk->sk_state = TCP_LISTEN; err = 0; @@ -1546,7 +1554,7 @@ static int vsock_stream_sendmsg(struct socket *sock, struct msghdr *msg, /* Callers should not provide a destination with stream sockets. */ if (msg->msg_namelen) { - err = sk->sk_state == SS_CONNECTED ? -EISCONN : -EOPNOTSUPP; + err = sk->sk_state == TCP_ESTABLISHED ? -EISCONN : -EOPNOTSUPP; goto out; } @@ -1557,7 +1565,7 @@ static int vsock_stream_sendmsg(struct socket *sock, struct msghdr *msg, goto out; } - if (sk->sk_state != SS_CONNECTED || + if (sk->sk_state != TCP_ESTABLISHED || !vsock_addr_bound(&vsk->local_addr)) { err = -ENOTCONN; goto out; @@ -1681,7 +1689,7 @@ vsock_stream_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, lock_sock(sk); - if (sk->sk_state != SS_CONNECTED) { + if (sk->sk_state != TCP_ESTABLISHED) { /* Recvmsg is supposed to return 0 if a peer performs an * orderly shutdown. Differentiate between that case and when a * peer has not connected or a local shutdown occured with the diff --git a/net/vmw_vsock/hyperv_transport.c b/net/vmw_vsock/hyperv_transport.c index 14ed5a344cdf..bbac023e70d1 100644 --- a/net/vmw_vsock/hyperv_transport.c +++ b/net/vmw_vsock/hyperv_transport.c @@ -310,7 +310,7 @@ static void hvs_close_connection(struct vmbus_channel *chan) struct sock *sk = get_per_channel_state(chan); struct vsock_sock *vsk = vsock_sk(sk); - sk->sk_state = SS_UNCONNECTED; + sk->sk_state = TCP_CLOSE; sock_set_flag(sk, SOCK_DONE); vsk->peer_shutdown |= SEND_SHUTDOWN | RCV_SHUTDOWN; @@ -344,8 +344,8 @@ static void hvs_open_connection(struct vmbus_channel *chan) if (!sk) return; - if ((conn_from_host && sk->sk_state != VSOCK_SS_LISTEN) || - (!conn_from_host && sk->sk_state != SS_CONNECTING)) + if ((conn_from_host && sk->sk_state != TCP_LISTEN) || + (!conn_from_host && sk->sk_state != TCP_SYN_SENT)) goto out; if (conn_from_host) { @@ -357,7 +357,7 @@ static void hvs_open_connection(struct vmbus_channel *chan) if (!new) goto out; - new->sk_state = SS_CONNECTING; + new->sk_state = TCP_SYN_SENT; vnew = vsock_sk(new); hvs_new = vnew->trans; hvs_new->chan = chan; @@ -384,7 +384,7 @@ static void hvs_open_connection(struct vmbus_channel *chan) vmbus_set_chn_rescind_callback(chan, hvs_close_connection); if (conn_from_host) { - new->sk_state = SS_CONNECTED; + new->sk_state = TCP_ESTABLISHED; sk->sk_ack_backlog++; hvs_addr_init(&vnew->local_addr, if_type); @@ -399,7 +399,7 @@ static void hvs_open_connection(struct vmbus_channel *chan) vsock_enqueue_accept(sk, new); release_sock(sk); } else { - sk->sk_state = SS_CONNECTED; + sk->sk_state = TCP_ESTABLISHED; sk->sk_socket->state = SS_CONNECTED; vsock_insert_connected(vsock_sk(sk)); diff --git a/net/vmw_vsock/virtio_transport.c b/net/vmw_vsock/virtio_transport.c index 403d86e80162..8e03bd3f3668 100644 --- a/net/vmw_vsock/virtio_transport.c +++ b/net/vmw_vsock/virtio_transport.c @@ -414,7 +414,7 @@ static void virtio_vsock_event_fill(struct virtio_vsock *vsock) static void virtio_vsock_reset_sock(struct sock *sk) { lock_sock(sk); - sk->sk_state = SS_UNCONNECTED; + sk->sk_state = TCP_CLOSE; sk->sk_err = ECONNRESET; sk->sk_error_report(sk); release_sock(sk); diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index edba7ab97563..3ae3a33da70b 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -708,7 +708,7 @@ static void virtio_transport_do_close(struct vsock_sock *vsk, sock_set_flag(sk, SOCK_DONE); vsk->peer_shutdown = SHUTDOWN_MASK; if (vsock_stream_has_data(vsk) <= 0) - sk->sk_state = SS_DISCONNECTING; + sk->sk_state = TCP_CLOSING; sk->sk_state_change(sk); if (vsk->close_work_scheduled && @@ -748,8 +748,8 @@ static bool virtio_transport_close(struct vsock_sock *vsk) { struct sock *sk = &vsk->sk; - if (!(sk->sk_state == SS_CONNECTED || - sk->sk_state == SS_DISCONNECTING)) + if (!(sk->sk_state == TCP_ESTABLISHED || + sk->sk_state == TCP_CLOSING)) return true; /* Already received SHUTDOWN from peer, reply with RST */ @@ -801,7 +801,7 @@ virtio_transport_recv_connecting(struct sock *sk, switch (le16_to_cpu(pkt->hdr.op)) { case VIRTIO_VSOCK_OP_RESPONSE: - sk->sk_state = SS_CONNECTED; + sk->sk_state = TCP_ESTABLISHED; sk->sk_socket->state = SS_CONNECTED; vsock_insert_connected(vsk); sk->sk_state_change(sk); @@ -821,7 +821,7 @@ virtio_transport_recv_connecting(struct sock *sk, destroy: virtio_transport_reset(vsk, pkt); - sk->sk_state = SS_UNCONNECTED; + sk->sk_state = TCP_CLOSE; sk->sk_err = skerr; sk->sk_error_report(sk); return err; @@ -857,7 +857,7 @@ virtio_transport_recv_connected(struct sock *sk, vsk->peer_shutdown |= SEND_SHUTDOWN; if (vsk->peer_shutdown == SHUTDOWN_MASK && vsock_stream_has_data(vsk) <= 0) - sk->sk_state = SS_DISCONNECTING; + sk->sk_state = TCP_CLOSING; if (le32_to_cpu(pkt->hdr.flags)) sk->sk_state_change(sk); break; @@ -928,7 +928,7 @@ virtio_transport_recv_listen(struct sock *sk, struct virtio_vsock_pkt *pkt) lock_sock_nested(child, SINGLE_DEPTH_NESTING); - child->sk_state = SS_CONNECTED; + child->sk_state = TCP_ESTABLISHED; vchild = vsock_sk(child); vsock_addr_init(&vchild->local_addr, le64_to_cpu(pkt->hdr.dst_cid), @@ -1016,18 +1016,18 @@ void virtio_transport_recv_pkt(struct virtio_vsock_pkt *pkt) sk->sk_write_space(sk); switch (sk->sk_state) { - case VSOCK_SS_LISTEN: + case TCP_LISTEN: virtio_transport_recv_listen(sk, pkt); virtio_transport_free_pkt(pkt); break; - case SS_CONNECTING: + case TCP_SYN_SENT: virtio_transport_recv_connecting(sk, pkt); virtio_transport_free_pkt(pkt); break; - case SS_CONNECTED: + case TCP_ESTABLISHED: virtio_transport_recv_connected(sk, pkt); break; - case SS_DISCONNECTING: + case TCP_CLOSING: virtio_transport_recv_disconnecting(sk, pkt); virtio_transport_free_pkt(pkt); break; diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c index 0206155bff53..391775e3575c 100644 --- a/net/vmw_vsock/vmci_transport.c +++ b/net/vmw_vsock/vmci_transport.c @@ -742,7 +742,7 @@ static int vmci_transport_recv_stream_cb(void *data, struct vmci_datagram *dg) /* The local context ID may be out of date, update it. */ vsk->local_addr.svm_cid = dst.svm_cid; - if (sk->sk_state == SS_CONNECTED) + if (sk->sk_state == TCP_ESTABLISHED) vmci_trans(vsk)->notify_ops->handle_notify_pkt( sk, pkt, true, &dst, &src, &bh_process_pkt); @@ -800,7 +800,9 @@ static void vmci_transport_handle_detach(struct sock *sk) * left in our consume queue. */ if (vsock_stream_has_data(vsk) <= 0) { - if (sk->sk_state == SS_CONNECTING) { + sk->sk_state = TCP_CLOSE; + + if (sk->sk_state == TCP_SYN_SENT) { /* The peer may detach from a queue pair while * we are still in the connecting state, i.e., * if the peer VM is killed after attaching to @@ -809,12 +811,10 @@ static void vmci_transport_handle_detach(struct sock *sk) * event like a reset. */ - sk->sk_state = SS_UNCONNECTED; sk->sk_err = ECONNRESET; sk->sk_error_report(sk); return; } - sk->sk_state = SS_UNCONNECTED; } sk->sk_state_change(sk); } @@ -882,17 +882,17 @@ static void vmci_transport_recv_pkt_work(struct work_struct *work) vsock_sk(sk)->local_addr.svm_cid = pkt->dg.dst.context; switch (sk->sk_state) { - case VSOCK_SS_LISTEN: + case TCP_LISTEN: vmci_transport_recv_listen(sk, pkt); break; - case SS_CONNECTING: + case TCP_SYN_SENT: /* Processing of pending connections for servers goes through * the listening socket, so see vmci_transport_recv_listen() * for that path. */ vmci_transport_recv_connecting_client(sk, pkt); break; - case SS_CONNECTED: + case TCP_ESTABLISHED: vmci_transport_recv_connected(sk, pkt); break; default: @@ -941,7 +941,7 @@ static int vmci_transport_recv_listen(struct sock *sk, vsock_sk(pending)->local_addr.svm_cid = pkt->dg.dst.context; switch (pending->sk_state) { - case SS_CONNECTING: + case TCP_SYN_SENT: err = vmci_transport_recv_connecting_server(sk, pending, pkt); @@ -1071,7 +1071,7 @@ static int vmci_transport_recv_listen(struct sock *sk, vsock_add_pending(sk, pending); sk->sk_ack_backlog++; - pending->sk_state = SS_CONNECTING; + pending->sk_state = TCP_SYN_SENT; vmci_trans(vpending)->produce_size = vmci_trans(vpending)->consume_size = qp_size; vmci_trans(vpending)->queue_pair_size = qp_size; @@ -1196,11 +1196,11 @@ vmci_transport_recv_connecting_server(struct sock *listener, * the socket will be valid until it is removed from the queue. * * If we fail sending the attach below, we remove the socket from the - * connected list and move the socket to SS_UNCONNECTED before + * connected list and move the socket to TCP_CLOSE before * releasing the lock, so a pending slow path processing of an incoming * packet will not see the socket in the connected state in that case. */ - pending->sk_state = SS_CONNECTED; + pending->sk_state = TCP_ESTABLISHED; vsock_insert_connected(vpending); @@ -1231,7 +1231,7 @@ vmci_transport_recv_connecting_server(struct sock *listener, destroy: pending->sk_err = skerr; - pending->sk_state = SS_UNCONNECTED; + pending->sk_state = TCP_CLOSE; /* As long as we drop our reference, all necessary cleanup will handle * when the cleanup function drops its reference and our destruct * implementation is called. Note that since the listen handler will @@ -1269,7 +1269,7 @@ vmci_transport_recv_connecting_client(struct sock *sk, * accounting (it can already be found since it's in the bound * table). */ - sk->sk_state = SS_CONNECTED; + sk->sk_state = TCP_ESTABLISHED; sk->sk_socket->state = SS_CONNECTED; vsock_insert_connected(vsk); sk->sk_state_change(sk); @@ -1337,7 +1337,7 @@ vmci_transport_recv_connecting_client(struct sock *sk, destroy: vmci_transport_send_reset(sk, pkt); - sk->sk_state = SS_UNCONNECTED; + sk->sk_state = TCP_CLOSE; sk->sk_err = skerr; sk->sk_error_report(sk); return err; @@ -1525,7 +1525,7 @@ static int vmci_transport_recv_connected(struct sock *sk, sock_set_flag(sk, SOCK_DONE); vsk->peer_shutdown = SHUTDOWN_MASK; if (vsock_stream_has_data(vsk) <= 0) - sk->sk_state = SS_DISCONNECTING; + sk->sk_state = TCP_CLOSING; sk->sk_state_change(sk); break; @@ -1789,7 +1789,7 @@ static int vmci_transport_connect(struct vsock_sock *vsk) err = vmci_transport_send_conn_request( sk, vmci_trans(vsk)->queue_pair_size); if (err < 0) { - sk->sk_state = SS_UNCONNECTED; + sk->sk_state = TCP_CLOSE; return err; } } else { @@ -1799,7 +1799,7 @@ static int vmci_transport_connect(struct vsock_sock *vsk) sk, vmci_trans(vsk)->queue_pair_size, supported_proto_versions); if (err < 0) { - sk->sk_state = SS_UNCONNECTED; + sk->sk_state = TCP_CLOSE; return err; } diff --git a/net/vmw_vsock/vmci_transport_notify.c b/net/vmw_vsock/vmci_transport_notify.c index 1406db4d97d1..41fb427f150a 100644 --- a/net/vmw_vsock/vmci_transport_notify.c +++ b/net/vmw_vsock/vmci_transport_notify.c @@ -355,7 +355,7 @@ vmci_transport_notify_pkt_poll_in(struct sock *sk, * queue. Ask for notifications when there is something to * read. */ - if (sk->sk_state == SS_CONNECTED) { + if (sk->sk_state == TCP_ESTABLISHED) { if (!send_waiting_read(sk, 1)) return -1; diff --git a/net/vmw_vsock/vmci_transport_notify_qstate.c b/net/vmw_vsock/vmci_transport_notify_qstate.c index f3a0afc46208..0cc84f2bb05e 100644 --- a/net/vmw_vsock/vmci_transport_notify_qstate.c +++ b/net/vmw_vsock/vmci_transport_notify_qstate.c @@ -176,7 +176,7 @@ vmci_transport_notify_pkt_poll_in(struct sock *sk, * queue. Ask for notifications when there is something to * read. */ - if (sk->sk_state == SS_CONNECTED) + if (sk->sk_state == TCP_ESTABLISHED) vsock_block_update_write_window(sk); *data_ready_now = false; } -- cgit v1.2.3 From 27204aaa9dc67b833b77179fdac556288ec3a4bf Mon Sep 17 00:00:00 2001 From: Wei Wang Date: Wed, 4 Oct 2017 10:03:44 -0700 Subject: tcp: uniform the set up of sockets after successful connection Currently in the TCP code, the initialization sequence for cached metrics, congestion control, BPF, etc, after successful connection is very inconsistent. This introduces inconsistent bevhavior and is prone to bugs. The current call sequence is as follows: (1) for active case (tcp_finish_connect() case): tcp_mtup_init(sk); icsk->icsk_af_ops->rebuild_header(sk); tcp_init_metrics(sk); tcp_call_bpf(sk, BPF_SOCK_OPS_ACTIVE_ESTABLISHED_CB); tcp_init_congestion_control(sk); tcp_init_buffer_space(sk); (2) for passive case (tcp_rcv_state_process() TCP_SYN_RECV case): icsk->icsk_af_ops->rebuild_header(sk); tcp_call_bpf(sk, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB); tcp_init_congestion_control(sk); tcp_mtup_init(sk); tcp_init_buffer_space(sk); tcp_init_metrics(sk); (3) for TFO passive case (tcp_fastopen_create_child()): inet_csk(child)->icsk_af_ops->rebuild_header(child); tcp_init_congestion_control(child); tcp_mtup_init(child); tcp_init_metrics(child); tcp_call_bpf(child, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB); tcp_init_buffer_space(child); This commit uniforms the above functions to have the following sequence: tcp_mtup_init(sk); icsk->icsk_af_ops->rebuild_header(sk); tcp_init_metrics(sk); tcp_call_bpf(sk, BPF_SOCK_OPS_ACTIVE/PASSIVE_ESTABLISHED_CB); tcp_init_congestion_control(sk); tcp_init_buffer_space(sk); This sequence is the same as the (1) active case. We pick this sequence because this order correctly allows BPF to override the settings including congestion control module and initial cwnd, etc from the route, and then allows the CC module to see those settings. Suggested-by: Neal Cardwell Tested-by: Neal Cardwell Signed-off-by: Wei Wang Acked-by: Neal Cardwell Acked-by: Yuchung Cheng Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/tcp.h | 1 + net/ipv4/tcp.c | 12 ++++++++++++ net/ipv4/tcp_fastopen.c | 7 +------ net/ipv4/tcp_input.c | 21 +++------------------ 4 files changed, 17 insertions(+), 24 deletions(-) (limited to 'include/net') diff --git a/include/net/tcp.h b/include/net/tcp.h index 7a3a8af56fd6..426c2e986016 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -416,6 +416,7 @@ bool tcp_peer_is_proven(struct request_sock *req, struct dst_entry *dst); void tcp_disable_fack(struct tcp_sock *tp); void tcp_close(struct sock *sk, long timeout); void tcp_init_sock(struct sock *sk); +void tcp_init_transfer(struct sock *sk, int bpf_op); unsigned int tcp_poll(struct file *file, struct socket *sock, struct poll_table_struct *wait); int tcp_getsockopt(struct sock *sk, int level, int optname, diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 23225c98d287..c115e37ca608 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -456,6 +456,18 @@ void tcp_init_sock(struct sock *sk) } EXPORT_SYMBOL(tcp_init_sock); +void tcp_init_transfer(struct sock *sk, int bpf_op) +{ + struct inet_connection_sock *icsk = inet_csk(sk); + + tcp_mtup_init(sk); + icsk->icsk_af_ops->rebuild_header(sk); + tcp_init_metrics(sk); + tcp_call_bpf(sk, bpf_op); + tcp_init_congestion_control(sk); + tcp_init_buffer_space(sk); +} + static void tcp_tx_timestamp(struct sock *sk, u16 tsflags, struct sk_buff *skb) { if (tsflags && skb) { diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index de470e7e586f..29fff14d5a53 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -236,12 +236,7 @@ static struct sock *tcp_fastopen_create_child(struct sock *sk, refcount_set(&req->rsk_refcnt, 2); /* Now finish processing the fastopen child socket. */ - inet_csk(child)->icsk_af_ops->rebuild_header(child); - tcp_init_congestion_control(child); - tcp_mtup_init(child); - tcp_init_metrics(child); - tcp_call_bpf(child, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB); - tcp_init_buffer_space(child); + tcp_init_transfer(child, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB); tp->rcv_nxt = TCP_SKB_CB(skb)->seq + 1; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index db9bb46b5776..bd3a35f5dbf2 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5513,20 +5513,13 @@ void tcp_finish_connect(struct sock *sk, struct sk_buff *skb) security_inet_conn_established(sk, skb); } - /* Make sure socket is routed, for correct metrics. */ - icsk->icsk_af_ops->rebuild_header(sk); - - tcp_init_metrics(sk); - tcp_call_bpf(sk, BPF_SOCK_OPS_ACTIVE_ESTABLISHED_CB); - tcp_init_congestion_control(sk); + tcp_init_transfer(sk, BPF_SOCK_OPS_ACTIVE_ESTABLISHED_CB); /* Prevent spurious tcp_cwnd_restart() on first data * packet. */ tp->lsndtime = tcp_jiffies32; - tcp_init_buffer_space(sk); - if (sock_flag(sk, SOCK_KEEPOPEN)) inet_csk_reset_keepalive_timer(sk, keepalive_time_when(tp)); @@ -5693,7 +5686,6 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, if (tcp_is_sack(tp) && sysctl_tcp_fack) tcp_enable_fack(tp); - tcp_mtup_init(sk); tcp_sync_mss(sk, icsk->icsk_pmtu_cookie); tcp_initialize_rcv_mss(sk); @@ -5920,14 +5912,8 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) inet_csk(sk)->icsk_retransmits = 0; reqsk_fastopen_remove(sk, req, false); } else { - /* Make sure socket is routed, for correct metrics. */ - icsk->icsk_af_ops->rebuild_header(sk); - tcp_call_bpf(sk, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB); - tcp_init_congestion_control(sk); - - tcp_mtup_init(sk); + tcp_init_transfer(sk, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB); tp->copied_seq = tp->rcv_nxt; - tcp_init_buffer_space(sk); } smp_mb(); tcp_set_state(sk, TCP_ESTABLISHED); @@ -5957,8 +5943,7 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) * are sent out. */ tcp_rearm_rto(sk); - } else - tcp_init_metrics(sk); + } if (!inet_csk(sk)->icsk_ca_ops->cong_control) tcp_update_pacing_rate(sk); -- cgit v1.2.3 From e2080072ed2d98a55ae69d95dea60ff7a17cddd5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 4 Oct 2017 12:59:58 -0700 Subject: tcp: new list for sent but unacked skbs for RACK recovery This patch adds a new queue (list) that tracks the sent but not yet acked or SACKed skbs for a TCP connection. The list is chronologically ordered by skb->skb_mstamp (the head is the oldest sent skb). This list will be used to optimize TCP Rack recovery, which checks an skb's timestamp to judge if it has been lost and needs to be retransmitted. Since TCP write queue is ordered by sequence instead of sent time, RACK has to scan over the write queue to catch all eligible packets to detect lost retransmission, and iterates through SACKed skbs repeatedly. Special cares for rare events: 1. TCP repair fakes skb transmission so the send queue needs adjusted 2. SACK reneging would require re-inserting SACKed skbs into the send queue. For now I believe it's not worth the complexity to make RACK work perfectly on SACK reneging, so we do nothing here. 3. Fast Open: currently for non-TFO, send-queue correctly queues the pure SYN packet. For TFO which queues a pure SYN and then a data packet, send-queue only queues the data packet but not the pure SYN due to the structure of TFO code. This is okay because the SYN receiver would never respond with a SACK on a missing SYN (i.e. SYN is never fast-retransmitted by SACK/RACK). In order to not grow sk_buff, we use an union for the new list and _skb_refdst/destructor fields. This is a bit complicated because we need to make sure _skb_refdst and destructor are properly zeroed before skb is cloned/copied at transmit, and before being freed. Signed-off-by: Eric Dumazet Signed-off-by: Yuchung Cheng Signed-off-by: Neal Cardwell Signed-off-by: David S. Miller --- include/linux/skbuff.h | 11 +++++++++-- include/linux/tcp.h | 1 + include/net/tcp.h | 24 +++++++++++++++++++++++- net/ipv4/tcp.c | 2 ++ net/ipv4/tcp_input.c | 9 +++++++-- net/ipv4/tcp_minisocks.c | 1 + net/ipv4/tcp_output.c | 42 +++++++++++++++++++++++++++++++----------- 7 files changed, 74 insertions(+), 16 deletions(-) (limited to 'include/net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index ada821466e88..01a985937867 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -617,6 +617,7 @@ typedef unsigned char *sk_buff_data_t; * @nf_trace: netfilter packet trace flag * @protocol: Packet protocol from driver * @destructor: Destruct function + * @tcp_tsorted_anchor: list structure for TCP (tp->tsorted_sent_queue) * @_nfct: Associated connection, if any (with nfctinfo bits) * @nf_bridge: Saved data about a bridged frame - see br_netfilter.c * @skb_iif: ifindex of device we arrived on @@ -686,8 +687,14 @@ struct sk_buff { */ char cb[48] __aligned(8); - unsigned long _skb_refdst; - void (*destructor)(struct sk_buff *skb); + union { + struct { + unsigned long _skb_refdst; + void (*destructor)(struct sk_buff *skb); + }; + struct list_head tcp_tsorted_anchor; + }; + #ifdef CONFIG_XFRM struct sec_path *sp; #endif diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 4aa40ef02d32..1d2c44e09e31 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -191,6 +191,7 @@ struct tcp_sock { u32 tsoffset; /* timestamp offset */ struct list_head tsq_node; /* anchor in tsq_tasklet.head list */ + struct list_head tsorted_sent_queue; /* time-sorted sent but un-SACKed skbs */ u32 snd_wl1; /* Sequence for window update */ u32 snd_wnd; /* The window we expect to receive */ diff --git a/include/net/tcp.h b/include/net/tcp.h index 426c2e986016..3b16f353b539 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1589,14 +1589,34 @@ enum tcp_chrono { void tcp_chrono_start(struct sock *sk, const enum tcp_chrono type); void tcp_chrono_stop(struct sock *sk, const enum tcp_chrono type); +/* This helper is needed, because skb->tcp_tsorted_anchor uses + * the same memory storage than skb->destructor/_skb_refdst + */ +static inline void tcp_skb_tsorted_anchor_cleanup(struct sk_buff *skb) +{ + skb->destructor = NULL; + skb->_skb_refdst = 0UL; +} + +#define tcp_skb_tsorted_save(skb) { \ + unsigned long _save = skb->_skb_refdst; \ + skb->_skb_refdst = 0UL; + +#define tcp_skb_tsorted_restore(skb) \ + skb->_skb_refdst = _save; \ +} + /* write queue abstraction */ static inline void tcp_write_queue_purge(struct sock *sk) { struct sk_buff *skb; tcp_chrono_stop(sk, TCP_CHRONO_BUSY); - while ((skb = __skb_dequeue(&sk->sk_write_queue)) != NULL) + while ((skb = __skb_dequeue(&sk->sk_write_queue)) != NULL) { + tcp_skb_tsorted_anchor_cleanup(skb); sk_wmem_free_skb(sk, skb); + } + INIT_LIST_HEAD(&tcp_sk(sk)->tsorted_sent_queue); sk_mem_reclaim(sk); tcp_clear_all_retrans_hints(tcp_sk(sk)); } @@ -1711,6 +1731,8 @@ static inline void tcp_insert_write_queue_before(struct sk_buff *new, static inline void tcp_unlink_write_queue(struct sk_buff *skb, struct sock *sk) { + list_del(&skb->tcp_tsorted_anchor); + tcp_skb_tsorted_anchor_cleanup(skb); __skb_unlink(skb, &sk->sk_write_queue); } diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index c115e37ca608..8cf742fd4f99 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -415,6 +415,7 @@ void tcp_init_sock(struct sock *sk) tp->out_of_order_queue = RB_ROOT; tcp_init_xmit_timers(sk); INIT_LIST_HEAD(&tp->tsq_node); + INIT_LIST_HEAD(&tp->tsorted_sent_queue); icsk->icsk_rto = TCP_TIMEOUT_INIT; tp->mdev_us = jiffies_to_usecs(TCP_TIMEOUT_INIT); @@ -881,6 +882,7 @@ struct sk_buff *sk_stream_alloc_skb(struct sock *sk, int size, gfp_t gfp, * available to the caller, no more, no less. */ skb->reserved_tailroom = skb->end - skb->tail - size; + INIT_LIST_HEAD(&skb->tcp_tsorted_anchor); return skb; } __kfree_skb(skb); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index c5b8d61846c2..fb0d7ed84b94 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -1593,6 +1593,8 @@ static struct sk_buff *tcp_sacktag_walk(struct sk_buff *skb, struct sock *sk, tcp_skb_pcount(skb), skb->skb_mstamp); tcp_rate_skb_delivered(sk, skb, state->rate); + if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED) + list_del_init(&skb->tcp_tsorted_anchor); if (!before(TCP_SKB_CB(skb)->seq, tcp_highest_sack_seq(tp))) @@ -3054,8 +3056,11 @@ static void tcp_ack_tstamp(struct sock *sk, struct sk_buff *skb, shinfo = skb_shinfo(skb); if (!before(shinfo->tskey, prior_snd_una) && - before(shinfo->tskey, tcp_sk(sk)->snd_una)) - __skb_tstamp_tx(skb, NULL, sk, SCM_TSTAMP_ACK); + before(shinfo->tskey, tcp_sk(sk)->snd_una)) { + tcp_skb_tsorted_save(skb) { + __skb_tstamp_tx(skb, NULL, sk, SCM_TSTAMP_ACK); + } tcp_skb_tsorted_restore(skb); + } } /* Remove acknowledged frames from the retransmission queue. If our packet diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 188a6f31356d..2341b9f857b6 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -446,6 +446,7 @@ struct sock *tcp_create_openreq_child(const struct sock *sk, newtp->snd_nxt = newtp->snd_up = treq->snt_isn + 1; INIT_LIST_HEAD(&newtp->tsq_node); + INIT_LIST_HEAD(&newtp->tsorted_sent_queue); tcp_init_wl(newtp, treq->rcv_isn); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 0bc9e46a5369..8162e2880178 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -971,6 +971,12 @@ static void tcp_internal_pacing(struct sock *sk, const struct sk_buff *skb) HRTIMER_MODE_ABS_PINNED); } +static void tcp_update_skb_after_send(struct tcp_sock *tp, struct sk_buff *skb) +{ + skb->skb_mstamp = tp->tcp_mstamp; + list_move_tail(&skb->tcp_tsorted_anchor, &tp->tsorted_sent_queue); +} + /* This routine actually transmits TCP packets queued in by * tcp_do_sendmsg(). This is used by both the initial * transmission and possible later retransmissions. @@ -1003,10 +1009,14 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, TCP_SKB_CB(skb)->tx.in_flight = TCP_SKB_CB(skb)->end_seq - tp->snd_una; oskb = skb; - if (unlikely(skb_cloned(skb))) - skb = pskb_copy(skb, gfp_mask); - else - skb = skb_clone(skb, gfp_mask); + + tcp_skb_tsorted_save(oskb) { + if (unlikely(skb_cloned(oskb))) + skb = pskb_copy(oskb, gfp_mask); + else + skb = skb_clone(oskb, gfp_mask); + } tcp_skb_tsorted_restore(oskb); + if (unlikely(!skb)) return -ENOBUFS; } @@ -1127,7 +1137,7 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, err = net_xmit_eval(err); } if (!err && oskb) { - oskb->skb_mstamp = tp->tcp_mstamp; + tcp_update_skb_after_send(tp, oskb); tcp_rate_skb_sent(sk, oskb); } return err; @@ -1328,6 +1338,7 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, /* Link BUFF into the send queue. */ __skb_header_release(buff); tcp_insert_write_queue_after(skb, buff, sk); + list_add(&buff->tcp_tsorted_anchor, &skb->tcp_tsorted_anchor); return 0; } @@ -2260,7 +2271,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, if (unlikely(tp->repair) && tp->repair_queue == TCP_SEND_QUEUE) { /* "skb_mstamp" is used as a start point for the retransmit timer */ - skb->skb_mstamp = tp->tcp_mstamp; + tcp_update_skb_after_send(tp, skb); goto repair; /* Skip network transmission */ } @@ -2838,11 +2849,14 @@ int __tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb, int segs) skb_headroom(skb) >= 0xFFFF)) { struct sk_buff *nskb; - nskb = __pskb_copy(skb, MAX_TCP_HEADER, GFP_ATOMIC); - err = nskb ? tcp_transmit_skb(sk, nskb, 0, GFP_ATOMIC) : - -ENOBUFS; + tcp_skb_tsorted_save(skb) { + nskb = __pskb_copy(skb, MAX_TCP_HEADER, GFP_ATOMIC); + err = nskb ? tcp_transmit_skb(sk, nskb, 0, GFP_ATOMIC) : + -ENOBUFS; + } tcp_skb_tsorted_restore(skb); + if (!err) - skb->skb_mstamp = tp->tcp_mstamp; + tcp_update_skb_after_send(tp, skb); } else { err = tcp_transmit_skb(sk, skb, 1, GFP_ATOMIC); } @@ -3023,6 +3037,7 @@ coalesce: goto coalesce; return; } + INIT_LIST_HEAD(&skb->tcp_tsorted_anchor); skb_reserve(skb, MAX_TCP_HEADER); sk_forced_mem_schedule(sk, skb->truesize); /* FIN eats a sequence byte, write_seq advanced by tcp_queue_skb(). */ @@ -3078,9 +3093,14 @@ int tcp_send_synack(struct sock *sk) } if (!(TCP_SKB_CB(skb)->tcp_flags & TCPHDR_ACK)) { if (skb_cloned(skb)) { - struct sk_buff *nskb = skb_copy(skb, GFP_ATOMIC); + struct sk_buff *nskb; + + tcp_skb_tsorted_save(skb) { + nskb = skb_copy(skb, GFP_ATOMIC); + } tcp_skb_tsorted_restore(skb); if (!nskb) return -ENOMEM; + INIT_LIST_HEAD(&nskb->tcp_tsorted_anchor); tcp_unlink_write_queue(skb, sk); __skb_header_release(nskb); __tcp_add_write_queue_head(sk, nskb); -- cgit v1.2.3 From 4e64b1ed15e25b8dcc2819c6d43dab72eb0bea26 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Thu, 5 Oct 2017 23:46:14 -0700 Subject: net/ipv6: Convert icmpv6_push_pending_frames to void commit cc71b7b07119 ("net/ipv6: remove unused err variable on icmpv6_push_pending_frames") exposed icmpv6_push_pending_frames return value not being used. Remove now unnecessary int err declarations and uses. Miscellanea: o Remove unnecessary goto and out: labels o Realign arguments Signed-off-by: Joe Perches Signed-off-by: David S. Miller --- include/net/ipv6.h | 4 ++-- net/ipv6/icmp.c | 43 ++++++++++++++++++------------------------- net/ipv6/ping.c | 5 ++--- 3 files changed, 22 insertions(+), 30 deletions(-) (limited to 'include/net') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 6eac5cf8f1e6..3cda3b521c36 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -300,8 +300,8 @@ static inline void fl6_sock_release(struct ip6_flowlabel *fl) void icmpv6_notify(struct sk_buff *skb, u8 type, u8 code, __be32 info); -int icmpv6_push_pending_frames(struct sock *sk, struct flowi6 *fl6, - struct icmp6hdr *thdr, int len); +void icmpv6_push_pending_frames(struct sock *sk, struct flowi6 *fl6, + struct icmp6hdr *thdr, int len); int ip6_ra_control(struct sock *sk, int sel); diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index aeb49b4d8c7d..4e52d52a6752 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -250,15 +250,15 @@ static bool opt_unrec(struct sk_buff *skb, __u32 offset) return (*op & 0xC0) == 0x80; } -int icmpv6_push_pending_frames(struct sock *sk, struct flowi6 *fl6, - struct icmp6hdr *thdr, int len) +void icmpv6_push_pending_frames(struct sock *sk, struct flowi6 *fl6, + struct icmp6hdr *thdr, int len) { struct sk_buff *skb; struct icmp6hdr *icmp6h; skb = skb_peek(&sk->sk_write_queue); if (!skb) - goto out; + return; icmp6h = icmp6_hdr(skb); memcpy(icmp6h, thdr, sizeof(struct icmp6hdr)); @@ -286,8 +286,6 @@ int icmpv6_push_pending_frames(struct sock *sk, struct flowi6 *fl6, tmp_csum); } ip6_push_pending_frames(sk); -out: - return 0; } struct icmpv6_msg { @@ -437,7 +435,6 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, int iif = 0; int addr_type = 0; int len; - int err = 0; u32 mark = IP6_REPLY_MARK(net, skb->mark); if ((u8 *)hdr < skb->head || @@ -574,17 +571,16 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, rcu_read_lock(); idev = __in6_dev_get(skb->dev); - err = ip6_append_data(sk, icmpv6_getfrag, &msg, - len + sizeof(struct icmp6hdr), - sizeof(struct icmp6hdr), - &ipc6, &fl6, (struct rt6_info *)dst, - MSG_DONTWAIT, &sockc_unused); - if (err) { + if (ip6_append_data(sk, icmpv6_getfrag, &msg, + len + sizeof(struct icmp6hdr), + sizeof(struct icmp6hdr), + &ipc6, &fl6, (struct rt6_info *)dst, + MSG_DONTWAIT, &sockc_unused)) { ICMP6_INC_STATS(net, idev, ICMP6_MIB_OUTERRORS); ip6_flush_pending_frames(sk); } else { - err = icmpv6_push_pending_frames(sk, &fl6, &tmp_hdr, - len + sizeof(struct icmp6hdr)); + icmpv6_push_pending_frames(sk, &fl6, &tmp_hdr, + len + sizeof(struct icmp6hdr)); } rcu_read_unlock(); out_dst_release: @@ -681,7 +677,6 @@ static void icmpv6_echo_reply(struct sk_buff *skb) struct icmpv6_msg msg; struct dst_entry *dst; struct ipcm6_cookie ipc6; - int err = 0; u32 mark = IP6_REPLY_MARK(net, skb->mark); struct sockcm_cookie sockc_unused = {0}; @@ -718,8 +713,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb) else if (!fl6.flowi6_oif) fl6.flowi6_oif = np->ucast_oif; - err = ip6_dst_lookup(net, sk, &dst, &fl6); - if (err) + if (ip6_dst_lookup(net, sk, &dst, &fl6)) goto out; dst = xfrm_lookup(net, dst, flowi6_to_flowi(&fl6), sk, 0); if (IS_ERR(dst)) @@ -736,17 +730,16 @@ static void icmpv6_echo_reply(struct sk_buff *skb) ipc6.dontfrag = np->dontfrag; ipc6.opt = NULL; - err = ip6_append_data(sk, icmpv6_getfrag, &msg, skb->len + sizeof(struct icmp6hdr), - sizeof(struct icmp6hdr), &ipc6, &fl6, - (struct rt6_info *)dst, MSG_DONTWAIT, - &sockc_unused); - - if (err) { + if (ip6_append_data(sk, icmpv6_getfrag, &msg, + skb->len + sizeof(struct icmp6hdr), + sizeof(struct icmp6hdr), &ipc6, &fl6, + (struct rt6_info *)dst, MSG_DONTWAIT, + &sockc_unused)) { __ICMP6_INC_STATS(net, idev, ICMP6_MIB_OUTERRORS); ip6_flush_pending_frames(sk); } else { - err = icmpv6_push_pending_frames(sk, &fl6, &tmp_hdr, - skb->len + sizeof(struct icmp6hdr)); + icmpv6_push_pending_frames(sk, &fl6, &tmp_hdr, + skb->len + sizeof(struct icmp6hdr)); } dst_release(dst); out: diff --git a/net/ipv6/ping.c b/net/ipv6/ping.c index ac826dd338ff..d12c55dad7d1 100644 --- a/net/ipv6/ping.c +++ b/net/ipv6/ping.c @@ -154,9 +154,8 @@ static int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) ICMP6_MIB_OUTERRORS); ip6_flush_pending_frames(sk); } else { - err = icmpv6_push_pending_frames(sk, &fl6, - (struct icmp6hdr *) &pfh.icmph, - len); + icmpv6_push_pending_frames(sk, &fl6, + (struct icmp6hdr *)&pfh.icmph, len); } release_sock(sk); -- cgit v1.2.3 From ac3f09ba3e496bd7cc780ead05b1d1bb5f33aedb Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 5 Oct 2017 22:21:22 -0700 Subject: tcp: uninline tcp_write_queue_purge() Since the upcoming rtx rbtree will add some extra code, it is time to not inline this fat function anymore. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/tcp.h | 15 +-------------- net/ipv4/tcp.c | 14 ++++++++++++++ 2 files changed, 15 insertions(+), 14 deletions(-) (limited to 'include/net') diff --git a/include/net/tcp.h b/include/net/tcp.h index 3b16f353b539..744559b72784 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1606,20 +1606,7 @@ static inline void tcp_skb_tsorted_anchor_cleanup(struct sk_buff *skb) skb->_skb_refdst = _save; \ } -/* write queue abstraction */ -static inline void tcp_write_queue_purge(struct sock *sk) -{ - struct sk_buff *skb; - - tcp_chrono_stop(sk, TCP_CHRONO_BUSY); - while ((skb = __skb_dequeue(&sk->sk_write_queue)) != NULL) { - tcp_skb_tsorted_anchor_cleanup(skb); - sk_wmem_free_skb(sk, skb); - } - INIT_LIST_HEAD(&tcp_sk(sk)->tsorted_sent_queue); - sk_mem_reclaim(sk); - tcp_clear_all_retrans_hints(tcp_sk(sk)); -} +void tcp_write_queue_purge(struct sock *sk); static inline struct sk_buff *tcp_write_queue_head(const struct sock *sk) { diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 8cf742fd4f99..f8ebae62f834 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2318,6 +2318,20 @@ static inline bool tcp_need_reset(int state) TCPF_FIN_WAIT2 | TCPF_SYN_RECV); } +void tcp_write_queue_purge(struct sock *sk) +{ + struct sk_buff *skb; + + tcp_chrono_stop(sk, TCP_CHRONO_BUSY); + while ((skb = __skb_dequeue(&sk->sk_write_queue)) != NULL) { + tcp_skb_tsorted_anchor_cleanup(skb); + sk_wmem_free_skb(sk, skb); + } + INIT_LIST_HEAD(&tcp_sk(sk)->tsorted_sent_queue); + sk_mem_reclaim(sk); + tcp_clear_all_retrans_hints(tcp_sk(sk)); +} + int tcp_disconnect(struct sock *sk, int flags) { struct inet_sock *inet = inet_sk(sk); -- cgit v1.2.3 From 75c119afe14f74b4dd967d75ed9f57ab6c0ef045 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 5 Oct 2017 22:21:27 -0700 Subject: tcp: implement rb-tree based retransmit queue Using a linear list to store all skbs in write queue has been okay for quite a while : O(N) is not too bad when N < 500. Things get messy when N is the order of 100,000 : Modern TCP stacks want 10Gbit+ of throughput even with 200 ms RTT flows. 40 ns per cache line miss means a full scan can use 4 ms, blowing away CPU caches. SACK processing often can use various hints to avoid parsing whole retransmit queue. But with high packet losses and/or high reordering, hints no longer work. Sender has to process thousands of unfriendly SACK, accumulating a huge socket backlog, burning a cpu and massively dropping packets. Using an rb-tree for retransmit queue has been avoided for years because it added complexity and overhead, but now is the time to be more resistant and say no to quadratic behavior. 1) RTX queue is no longer part of the write queue : already sent skbs are stored in one rb-tree. 2) Since reaching the head of write queue no longer needs sk->sk_send_head, we added an union of sk_send_head and tcp_rtx_queue Tested: On receiver : netem on ingress : delay 150ms 200us loss 1 GRO disabled to force stress and SACK storms. for f in `seq 1 10` do ./netperf -H lpaa6 -l30 -- -K bbr -o THROUGHPUT|tail -1 done | awk '{print $0} {sum += $0} END {printf "%7u\n",sum}' Before patch : 323.87 351.48 339.59 338.62 306.72 204.07 304.93 291.88 202.47 176.88 2840 After patch: 1700.83 2207.98 2070.17 1544.26 2114.76 2124.89 1693.14 1080.91 2216.82 1299.94 18053 Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/sock.h | 7 ++- include/net/tcp.h | 89 ++++++++++++++++---------------- net/ipv4/tcp.c | 41 +++++++++++---- net/ipv4/tcp_input.c | 133 +++++++++++++++++++++++++----------------------- net/ipv4/tcp_ipv4.c | 2 +- net/ipv4/tcp_output.c | 137 +++++++++++++++++++++++++++----------------------- net/ipv4/tcp_timer.c | 24 +++++---- 7 files changed, 245 insertions(+), 188 deletions(-) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index a6b9a8d1a6df..4827094f1db4 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -60,7 +60,7 @@ #include #include #include - +#include #include #include #include @@ -397,7 +397,10 @@ struct sock { int sk_wmem_queued; refcount_t sk_wmem_alloc; unsigned long sk_tsq_flags; - struct sk_buff *sk_send_head; + union { + struct sk_buff *sk_send_head; + struct rb_root tcp_rtx_queue; + }; struct sk_buff_head sk_write_queue; __s32 sk_peek_off; int sk_write_pending; diff --git a/include/net/tcp.h b/include/net/tcp.h index 744559b72784..5a95e5886b55 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -551,7 +551,13 @@ void tcp_xmit_retransmit_queue(struct sock *); void tcp_simple_retransmit(struct sock *); void tcp_enter_recovery(struct sock *sk, bool ece_ack); int tcp_trim_head(struct sock *, struct sk_buff *, u32); -int tcp_fragment(struct sock *, struct sk_buff *, u32, unsigned int, gfp_t); +enum tcp_queue { + TCP_FRAG_IN_WRITE_QUEUE, + TCP_FRAG_IN_RTX_QUEUE, +}; +int tcp_fragment(struct sock *sk, enum tcp_queue tcp_queue, + struct sk_buff *skb, u32 len, + unsigned int mss_now, gfp_t gfp); void tcp_send_probe0(struct sock *); void tcp_send_partial(struct sock *); @@ -1608,6 +1614,11 @@ static inline void tcp_skb_tsorted_anchor_cleanup(struct sk_buff *skb) void tcp_write_queue_purge(struct sock *sk); +static inline struct sk_buff *tcp_rtx_queue_head(const struct sock *sk) +{ + return skb_rb_first(&sk->tcp_rtx_queue); +} + static inline struct sk_buff *tcp_write_queue_head(const struct sock *sk) { return skb_peek(&sk->sk_write_queue); @@ -1630,18 +1641,12 @@ static inline struct sk_buff *tcp_write_queue_prev(const struct sock *sk, return skb_queue_prev(&sk->sk_write_queue, skb); } -#define tcp_for_write_queue(skb, sk) \ - skb_queue_walk(&(sk)->sk_write_queue, skb) - -#define tcp_for_write_queue_from(skb, sk) \ - skb_queue_walk_from(&(sk)->sk_write_queue, skb) - #define tcp_for_write_queue_from_safe(skb, tmp, sk) \ skb_queue_walk_from_safe(&(sk)->sk_write_queue, skb, tmp) static inline struct sk_buff *tcp_send_head(const struct sock *sk) { - return sk->sk_send_head; + return skb_peek(&sk->sk_write_queue); } static inline bool tcp_skb_is_last(const struct sock *sk, @@ -1650,29 +1655,30 @@ static inline bool tcp_skb_is_last(const struct sock *sk, return skb_queue_is_last(&sk->sk_write_queue, skb); } -static inline void tcp_advance_send_head(struct sock *sk, const struct sk_buff *skb) +static inline bool tcp_write_queue_empty(const struct sock *sk) { - if (tcp_skb_is_last(sk, skb)) - sk->sk_send_head = NULL; - else - sk->sk_send_head = tcp_write_queue_next(sk, skb); + return skb_queue_empty(&sk->sk_write_queue); +} + +static inline bool tcp_rtx_queue_empty(const struct sock *sk) +{ + return RB_EMPTY_ROOT(&sk->tcp_rtx_queue); +} + +static inline bool tcp_rtx_and_write_queues_empty(const struct sock *sk) +{ + return tcp_rtx_queue_empty(sk) && tcp_write_queue_empty(sk); } static inline void tcp_check_send_head(struct sock *sk, struct sk_buff *skb_unlinked) { - if (sk->sk_send_head == skb_unlinked) { - sk->sk_send_head = NULL; + if (tcp_write_queue_empty(sk)) tcp_chrono_stop(sk, TCP_CHRONO_BUSY); - } + if (tcp_sk(sk)->highest_sack == skb_unlinked) tcp_sk(sk)->highest_sack = NULL; } -static inline void tcp_init_send_head(struct sock *sk) -{ - sk->sk_send_head = NULL; -} - static inline void __tcp_add_write_queue_tail(struct sock *sk, struct sk_buff *skb) { __skb_queue_tail(&sk->sk_write_queue, skb); @@ -1683,8 +1689,7 @@ static inline void tcp_add_write_queue_tail(struct sock *sk, struct sk_buff *skb __tcp_add_write_queue_tail(sk, skb); /* Queue it, remembering where we must start sending. */ - if (sk->sk_send_head == NULL) { - sk->sk_send_head = skb; + if (sk->sk_write_queue.next == skb) { tcp_chrono_start(sk, TCP_CHRONO_BUSY); if (tcp_sk(sk)->highest_sack == NULL) @@ -1697,35 +1702,32 @@ static inline void __tcp_add_write_queue_head(struct sock *sk, struct sk_buff *s __skb_queue_head(&sk->sk_write_queue, skb); } -/* Insert buff after skb on the write queue of sk. */ -static inline void tcp_insert_write_queue_after(struct sk_buff *skb, - struct sk_buff *buff, - struct sock *sk) -{ - __skb_queue_after(&sk->sk_write_queue, skb, buff); -} - /* Insert new before skb on the write queue of sk. */ static inline void tcp_insert_write_queue_before(struct sk_buff *new, struct sk_buff *skb, struct sock *sk) { __skb_queue_before(&sk->sk_write_queue, skb, new); - - if (sk->sk_send_head == skb) - sk->sk_send_head = new; } static inline void tcp_unlink_write_queue(struct sk_buff *skb, struct sock *sk) { - list_del(&skb->tcp_tsorted_anchor); - tcp_skb_tsorted_anchor_cleanup(skb); __skb_unlink(skb, &sk->sk_write_queue); } -static inline bool tcp_write_queue_empty(struct sock *sk) +void tcp_rbtree_insert(struct rb_root *root, struct sk_buff *skb); + +static inline void tcp_rtx_queue_unlink(struct sk_buff *skb, struct sock *sk) { - return skb_queue_empty(&sk->sk_write_queue); + tcp_skb_tsorted_anchor_cleanup(skb); + rb_erase(&skb->rbnode, &sk->tcp_rtx_queue); +} + +static inline void tcp_rtx_queue_unlink_and_free(struct sk_buff *skb, struct sock *sk) +{ + list_del(&skb->tcp_tsorted_anchor); + tcp_rtx_queue_unlink(skb, sk); + sk_wmem_free_skb(sk, skb); } static inline void tcp_push_pending_frames(struct sock *sk) @@ -1754,8 +1756,9 @@ static inline u32 tcp_highest_sack_seq(struct tcp_sock *tp) static inline void tcp_advance_highest_sack(struct sock *sk, struct sk_buff *skb) { - tcp_sk(sk)->highest_sack = tcp_skb_is_last(sk, skb) ? NULL : - tcp_write_queue_next(sk, skb); + struct sk_buff *next = skb_rb_next(skb); + + tcp_sk(sk)->highest_sack = next ?: tcp_send_head(sk); } static inline struct sk_buff *tcp_highest_sack(struct sock *sk) @@ -1765,7 +1768,9 @@ static inline struct sk_buff *tcp_highest_sack(struct sock *sk) static inline void tcp_highest_sack_reset(struct sock *sk) { - tcp_sk(sk)->highest_sack = tcp_write_queue_head(sk); + struct sk_buff *skb = tcp_rtx_queue_head(sk); + + tcp_sk(sk)->highest_sack = skb ?: tcp_send_head(sk); } /* Called when old skb is about to be deleted (to be combined with new skb) */ @@ -1935,7 +1940,7 @@ extern void tcp_rack_reo_timeout(struct sock *sk); /* At how many usecs into the future should the RTO fire? */ static inline s64 tcp_rto_delta_us(const struct sock *sk) { - const struct sk_buff *skb = tcp_write_queue_head(sk); + const struct sk_buff *skb = tcp_rtx_queue_head(sk); u32 rto = inet_csk(sk)->icsk_rto; u64 rto_time_stamp_us = skb->skb_mstamp + jiffies_to_usecs(rto); diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index b8d379c80936..3b34850d361f 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -413,6 +413,7 @@ void tcp_init_sock(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); tp->out_of_order_queue = RB_ROOT; + sk->tcp_rtx_queue = RB_ROOT; tcp_init_xmit_timers(sk); INIT_LIST_HEAD(&tp->tsq_node); INIT_LIST_HEAD(&tp->tsorted_sent_queue); @@ -701,10 +702,9 @@ static void tcp_push(struct sock *sk, int flags, int mss_now, struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; - if (!tcp_send_head(sk)) - return; - skb = tcp_write_queue_tail(sk); + if (!skb) + return; if (!(flags & MSG_MORE) || forced_push(tp)) tcp_mark_push(tp, skb); @@ -964,14 +964,14 @@ ssize_t do_tcp_sendpages(struct sock *sk, struct page *page, int offset, int copy, i; bool can_coalesce; - if (!tcp_send_head(sk) || (copy = size_goal - skb->len) <= 0 || + if (!skb || (copy = size_goal - skb->len) <= 0 || !tcp_skb_can_collapse_to(skb)) { new_segment: if (!sk_stream_memory_free(sk)) goto wait_for_sndbuf; skb = sk_stream_alloc_skb(sk, 0, sk->sk_allocation, - skb_queue_empty(&sk->sk_write_queue)); + tcp_rtx_and_write_queues_empty(sk)); if (!skb) goto wait_for_memory; @@ -1199,7 +1199,7 @@ int tcp_sendmsg_locked(struct sock *sk, struct msghdr *msg, size_t size) goto out_err; } - skb = tcp_send_head(sk) ? tcp_write_queue_tail(sk) : NULL; + skb = tcp_write_queue_tail(sk); uarg = sock_zerocopy_realloc(sk, size, skb_zcopy(skb)); if (!uarg) { err = -ENOBUFS; @@ -1275,7 +1275,7 @@ restart: int max = size_goal; skb = tcp_write_queue_tail(sk); - if (tcp_send_head(sk)) { + if (skb) { if (skb->ip_summed == CHECKSUM_NONE) max = mss_now; copy = max - skb->len; @@ -1295,7 +1295,7 @@ new_segment: process_backlog = false; goto restart; } - first_skb = skb_queue_empty(&sk->sk_write_queue); + first_skb = tcp_rtx_and_write_queues_empty(sk); skb = sk_stream_alloc_skb(sk, select_size(sk, sg, first_skb), sk->sk_allocation, @@ -1521,6 +1521,13 @@ static int tcp_peek_sndq(struct sock *sk, struct msghdr *msg, int len) /* XXX -- need to support SO_PEEK_OFF */ + skb_rbtree_walk(skb, &sk->tcp_rtx_queue) { + err = skb_copy_datagram_msg(skb, 0, msg, skb->len); + if (err) + return err; + copied += skb->len; + } + skb_queue_walk(&sk->sk_write_queue, skb) { err = skb_copy_datagram_msg(skb, 0, msg, skb->len); if (err) @@ -2320,6 +2327,22 @@ static inline bool tcp_need_reset(int state) TCPF_FIN_WAIT2 | TCPF_SYN_RECV); } +static void tcp_rtx_queue_purge(struct sock *sk) +{ + struct rb_node *p = rb_first(&sk->tcp_rtx_queue); + + while (p) { + struct sk_buff *skb = rb_to_skb(p); + + p = rb_next(p); + /* Since we are deleting whole queue, no need to + * list_del(&skb->tcp_tsorted_anchor) + */ + tcp_rtx_queue_unlink(skb, sk); + sk_wmem_free_skb(sk, skb); + } +} + void tcp_write_queue_purge(struct sock *sk) { struct sk_buff *skb; @@ -2329,6 +2352,7 @@ void tcp_write_queue_purge(struct sock *sk) tcp_skb_tsorted_anchor_cleanup(skb); sk_wmem_free_skb(sk, skb); } + tcp_rtx_queue_purge(sk); INIT_LIST_HEAD(&tcp_sk(sk)->tsorted_sent_queue); sk_mem_reclaim(sk); tcp_clear_all_retrans_hints(tcp_sk(sk)); @@ -2392,7 +2416,6 @@ int tcp_disconnect(struct sock *sk, int flags) * issue in __tcp_select_window() */ icsk->icsk_ack.rcv_mss = TCP_MIN_MSS; - tcp_init_send_head(sk); memset(&tp->rx_opt, 0, sizeof(tp->rx_opt)); __sk_dst_reset(sk); dst_release(sk->sk_rx_dst); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 72c4732ae2da..d0682ce2a5d6 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -1142,6 +1142,7 @@ struct tcp_sacktag_state { u64 last_sackt; struct rate_sample *rate; int flag; + unsigned int mss_now; }; /* Check if skb is fully within the SACK block. In presence of GSO skbs, @@ -1191,7 +1192,8 @@ static int tcp_match_skb_to_sack(struct sock *sk, struct sk_buff *skb, if (pkt_len >= skb->len && !in_sack) return 0; - err = tcp_fragment(sk, skb, pkt_len, mss, GFP_ATOMIC); + err = tcp_fragment(sk, TCP_FRAG_IN_RTX_QUEUE, skb, + pkt_len, mss, GFP_ATOMIC); if (err < 0) return err; } @@ -1363,8 +1365,7 @@ static bool tcp_shifted_skb(struct sock *sk, struct sk_buff *prev, if (unlikely(TCP_SKB_CB(prev)->tx.delivered_mstamp)) TCP_SKB_CB(prev)->tx.delivered_mstamp = 0; - tcp_unlink_write_queue(skb, sk); - sk_wmem_free_skb(sk, skb); + tcp_rtx_queue_unlink_and_free(skb, sk); NET_INC_STATS(sock_net(sk), LINUX_MIB_SACKMERGED); @@ -1414,9 +1415,9 @@ static struct sk_buff *tcp_shift_skb_data(struct sock *sk, struct sk_buff *skb, goto fallback; /* Can only happen with delayed DSACK + discard craziness */ - if (unlikely(skb == tcp_write_queue_head(sk))) + prev = skb_rb_prev(skb); + if (!prev) goto fallback; - prev = tcp_write_queue_prev(sk, skb); if ((TCP_SKB_CB(prev)->sacked & TCPCB_TAGBITS) != TCPCB_SACKED_ACKED) goto fallback; @@ -1501,12 +1502,11 @@ static struct sk_buff *tcp_shift_skb_data(struct sock *sk, struct sk_buff *skb, /* Hole filled allows collapsing with the next as well, this is very * useful when hole on every nth skb pattern happens */ - if (prev == tcp_write_queue_tail(sk)) + skb = skb_rb_next(prev); + if (!skb) goto out; - skb = tcp_write_queue_next(sk, prev); if (!skb_can_shift(skb) || - (skb == tcp_send_head(sk)) || ((TCP_SKB_CB(skb)->sacked & TCPCB_TAGBITS) != TCPCB_SACKED_ACKED) || (mss != tcp_skb_seglen(skb))) goto out; @@ -1539,13 +1539,10 @@ static struct sk_buff *tcp_sacktag_walk(struct sk_buff *skb, struct sock *sk, struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *tmp; - tcp_for_write_queue_from(skb, sk) { + skb_rbtree_walk_from(skb) { int in_sack = 0; bool dup_sack = dup_sack_in; - if (skb == tcp_send_head(sk)) - break; - /* queue is in-order => we can short-circuit the walk early */ if (!before(TCP_SKB_CB(skb)->seq, end_seq)) break; @@ -1607,23 +1604,44 @@ static struct sk_buff *tcp_sacktag_walk(struct sk_buff *skb, struct sock *sk, return skb; } -/* Avoid all extra work that is being done by sacktag while walking in - * a normal way - */ +static struct sk_buff *tcp_sacktag_bsearch(struct sock *sk, + struct tcp_sacktag_state *state, + u32 seq) +{ + struct rb_node *parent, **p = &sk->tcp_rtx_queue.rb_node; + struct sk_buff *skb; + int unack_bytes; + + while (*p) { + parent = *p; + skb = rb_to_skb(parent); + if (before(seq, TCP_SKB_CB(skb)->seq)) { + p = &parent->rb_left; + continue; + } + if (!before(seq, TCP_SKB_CB(skb)->end_seq)) { + p = &parent->rb_right; + continue; + } + + state->fack_count = 0; + unack_bytes = TCP_SKB_CB(skb)->seq - tcp_sk(sk)->snd_una; + if (state->mss_now && unack_bytes > 0) + state->fack_count = unack_bytes / state->mss_now; + + return skb; + } + return NULL; +} + static struct sk_buff *tcp_sacktag_skip(struct sk_buff *skb, struct sock *sk, struct tcp_sacktag_state *state, u32 skip_to_seq) { - tcp_for_write_queue_from(skb, sk) { - if (skb == tcp_send_head(sk)) - break; - - if (after(TCP_SKB_CB(skb)->end_seq, skip_to_seq)) - break; + if (skb && after(TCP_SKB_CB(skb)->seq, skip_to_seq)) + return skb; - state->fack_count += tcp_skb_pcount(skb); - } - return skb; + return tcp_sacktag_bsearch(sk, state, skip_to_seq); } static struct sk_buff *tcp_maybe_skipping_dsack(struct sk_buff *skb, @@ -1745,8 +1763,9 @@ tcp_sacktag_write_queue(struct sock *sk, const struct sk_buff *ack_skb, } } - skb = tcp_write_queue_head(sk); + state->mss_now = tcp_current_mss(sk); state->fack_count = 0; + skb = NULL; i = 0; if (!tp->sacked_out) { @@ -1970,7 +1989,7 @@ void tcp_enter_loss(struct sock *sk) if (tcp_is_reno(tp)) tcp_reset_reno_sack(tp); - skb = tcp_write_queue_head(sk); + skb = tcp_rtx_queue_head(sk); is_reneg = skb && (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED); if (is_reneg) { NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPSACKRENEGING); @@ -1979,10 +1998,7 @@ void tcp_enter_loss(struct sock *sk) } tcp_clear_all_retrans_hints(tp); - tcp_for_write_queue(skb, sk) { - if (skb == tcp_send_head(sk)) - break; - + skb_rbtree_walk_from(skb) { mark_lost = (!(TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED) || is_reneg); if (mark_lost) @@ -2215,13 +2231,11 @@ static void tcp_mark_head_lost(struct sock *sk, int packets, int mark_head) return; cnt = tp->lost_cnt_hint; } else { - skb = tcp_write_queue_head(sk); + skb = tcp_rtx_queue_head(sk); cnt = 0; } - tcp_for_write_queue_from(skb, sk) { - if (skb == tcp_send_head(sk)) - break; + skb_rbtree_walk_from(skb) { /* TODO: do this better */ /* this is not the most efficient way to do this... */ tp->lost_skb_hint = skb; @@ -2245,7 +2259,8 @@ static void tcp_mark_head_lost(struct sock *sk, int packets, int mark_head) /* If needed, chop off the prefix to mark as lost. */ lost = (packets - oldcnt) * mss; if (lost < skb->len && - tcp_fragment(sk, skb, lost, mss, GFP_ATOMIC) < 0) + tcp_fragment(sk, TCP_FRAG_IN_RTX_QUEUE, skb, + lost, mss, GFP_ATOMIC) < 0) break; cnt = packets; } @@ -2329,7 +2344,7 @@ static bool tcp_any_retrans_done(const struct sock *sk) if (tp->retrans_out) return true; - skb = tcp_write_queue_head(sk); + skb = tcp_rtx_queue_head(sk); if (unlikely(skb && TCP_SKB_CB(skb)->sacked & TCPCB_EVER_RETRANS)) return true; @@ -2370,9 +2385,7 @@ static void tcp_undo_cwnd_reduction(struct sock *sk, bool unmark_loss) if (unmark_loss) { struct sk_buff *skb; - tcp_for_write_queue(skb, sk) { - if (skb == tcp_send_head(sk)) - break; + skb_rbtree_walk(skb, &sk->tcp_rtx_queue) { TCP_SKB_CB(skb)->sacked &= ~TCPCB_LOST; } tp->lost_out = 0; @@ -2617,9 +2630,7 @@ void tcp_simple_retransmit(struct sock *sk) unsigned int mss = tcp_current_mss(sk); u32 prior_lost = tp->lost_out; - tcp_for_write_queue(skb, sk) { - if (skb == tcp_send_head(sk)) - break; + skb_rbtree_walk(skb, &sk->tcp_rtx_queue) { if (tcp_skb_seglen(skb) > mss && !(TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) { if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_RETRANS) { @@ -2713,7 +2724,7 @@ static void tcp_process_loss(struct sock *sk, int flag, bool is_dupack, * is updated in tcp_ack()). Otherwise fall back to * the conventional recovery. */ - if (tcp_send_head(sk) && + if (!tcp_write_queue_empty(sk) && after(tcp_wnd_end(tp), tp->snd_nxt)) { *rexmit = REXMIT_NEW; return; @@ -3077,11 +3088,11 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, struct tcp_sock *tp = tcp_sk(sk); u32 prior_sacked = tp->sacked_out; u32 reord = tp->packets_out; + struct sk_buff *skb, *next; bool fully_acked = true; long sack_rtt_us = -1L; long seq_rtt_us = -1L; long ca_rtt_us = -1L; - struct sk_buff *skb; u32 pkts_acked = 0; u32 last_in_flight = 0; bool rtt_update; @@ -3089,7 +3100,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, first_ackt = 0; - while ((skb = tcp_write_queue_head(sk)) && skb != tcp_send_head(sk)) { + for (skb = skb_rb_first(&sk->tcp_rtx_queue); skb; skb = next) { struct tcp_skb_cb *scb = TCP_SKB_CB(skb); u8 sacked = scb->sacked; u32 acked_pcount; @@ -3107,8 +3118,6 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, break; fully_acked = false; } else { - /* Speedup tcp_unlink_write_queue() and next loop */ - prefetchw(skb->next); acked_pcount = tcp_skb_pcount(skb); } @@ -3160,12 +3169,12 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, if (!fully_acked) break; - tcp_unlink_write_queue(skb, sk); - sk_wmem_free_skb(sk, skb); + next = skb_rb_next(skb); if (unlikely(skb == tp->retransmit_skb_hint)) tp->retransmit_skb_hint = NULL; if (unlikely(skb == tp->lost_skb_hint)) tp->lost_skb_hint = NULL; + tcp_rtx_queue_unlink_and_free(skb, sk); } if (!skb) @@ -3257,12 +3266,14 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, static void tcp_ack_probe(struct sock *sk) { - const struct tcp_sock *tp = tcp_sk(sk); struct inet_connection_sock *icsk = inet_csk(sk); + struct sk_buff *head = tcp_send_head(sk); + const struct tcp_sock *tp = tcp_sk(sk); /* Was it a usable window open? */ - - if (!after(TCP_SKB_CB(tcp_send_head(sk))->end_seq, tcp_wnd_end(tp))) { + if (!head) + return; + if (!after(TCP_SKB_CB(head)->end_seq, tcp_wnd_end(tp))) { icsk->icsk_backoff = 0; inet_csk_clear_xmit_timer(sk, ICSK_TIME_PROBE0); /* Socket must be waked up by subsequent tcp_data_snd_check(). @@ -3382,7 +3393,7 @@ static int tcp_ack_update_window(struct sock *sk, const struct sk_buff *skb, u32 tp->pred_flags = 0; tcp_fast_path_check(sk); - if (tcp_send_head(sk)) + if (!tcp_write_queue_empty(sk)) tcp_slow_start_after_idle_check(sk); if (nwin > tp->max_window) { @@ -3567,8 +3578,8 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) sack_state.first_sackt = 0; sack_state.rate = &rs; - /* We very likely will need to access write queue head. */ - prefetchw(sk->sk_write_queue.next); + /* We very likely will need to access rtx queue. */ + prefetch(sk->tcp_rtx_queue.rb_node); /* If the ack is older than previous acks * then we can probably ignore it. @@ -3682,8 +3693,7 @@ no_queue: * being used to time the probes, and is probably far higher than * it needs to be for normal retransmission. */ - if (tcp_send_head(sk)) - tcp_ack_probe(sk); + tcp_ack_probe(sk); if (tp->tlp_high_seq) tcp_process_tlp_ack(sk, ack, flag); @@ -4726,7 +4736,7 @@ static struct sk_buff *tcp_collapse_one(struct sock *sk, struct sk_buff *skb, } /* Insert skb into rb tree, ordered by TCP_SKB_CB(skb)->seq */ -static void tcp_rbtree_insert(struct rb_root *root, struct sk_buff *skb) +void tcp_rbtree_insert(struct rb_root *root, struct sk_buff *skb) { struct rb_node **p = &root->rb_node; struct rb_node *parent = NULL; @@ -5530,7 +5540,7 @@ static bool tcp_rcv_fastopen_synack(struct sock *sk, struct sk_buff *synack, struct tcp_fastopen_cookie *cookie) { struct tcp_sock *tp = tcp_sk(sk); - struct sk_buff *data = tp->syn_data ? tcp_write_queue_head(sk) : NULL; + struct sk_buff *data = tp->syn_data ? tcp_rtx_queue_head(sk) : NULL; u16 mss = tp->rx_opt.mss_clamp, try_exp = 0; bool syn_drop = false; @@ -5565,9 +5575,8 @@ static bool tcp_rcv_fastopen_synack(struct sock *sk, struct sk_buff *synack, tcp_fastopen_cache_set(sk, mss, cookie, syn_drop, try_exp); if (data) { /* Retransmit unacked data in SYN */ - tcp_for_write_queue_from(data, sk) { - if (data == tcp_send_head(sk) || - __tcp_retransmit_skb(sk, data, 1)) + skb_rbtree_walk_from(data) { + if (__tcp_retransmit_skb(sk, data, 1)) break; } tcp_rearm_rto(sk); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index c7460fd90884..5418ecf03b78 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -480,7 +480,7 @@ void tcp_v4_err(struct sk_buff *icmp_skb, u32 info) TCP_TIMEOUT_INIT; icsk->icsk_rto = inet_csk_rto_backoff(icsk, TCP_RTO_MAX); - skb = tcp_write_queue_head(sk); + skb = tcp_rtx_queue_head(sk); BUG_ON(!skb); tcp_mstamp_refresh(tp); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 8162e2880178..696b0a168f16 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -66,15 +66,17 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, int push_one, gfp_t gfp); /* Account for new data that has been sent to the network. */ -static void tcp_event_new_data_sent(struct sock *sk, const struct sk_buff *skb) +static void tcp_event_new_data_sent(struct sock *sk, struct sk_buff *skb) { struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); unsigned int prior_packets = tp->packets_out; - tcp_advance_send_head(sk, skb); tp->snd_nxt = TCP_SKB_CB(skb)->end_seq; + __skb_unlink(skb, &sk->sk_write_queue); + tcp_rbtree_insert(&sk->tcp_rtx_queue, skb); + tp->packets_out += tcp_skb_pcount(skb); if (!prior_packets || icsk->icsk_pending == ICSK_TIME_LOSS_PROBE) tcp_rearm_rto(sk); @@ -1249,12 +1251,25 @@ static void tcp_skb_fragment_eor(struct sk_buff *skb, struct sk_buff *skb2) TCP_SKB_CB(skb)->eor = 0; } +/* Insert buff after skb on the write or rtx queue of sk. */ +static void tcp_insert_write_queue_after(struct sk_buff *skb, + struct sk_buff *buff, + struct sock *sk, + enum tcp_queue tcp_queue) +{ + if (tcp_queue == TCP_FRAG_IN_WRITE_QUEUE) + __skb_queue_after(&sk->sk_write_queue, skb, buff); + else + tcp_rbtree_insert(&sk->tcp_rtx_queue, buff); +} + /* Function to create two new TCP segments. Shrinks the given segment * to the specified size and appends a new segment with the rest of the * packet to the list. This won't be called frequently, I hope. * Remember, these are still headerless SKBs at this point. */ -int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, +int tcp_fragment(struct sock *sk, enum tcp_queue tcp_queue, + struct sk_buff *skb, u32 len, unsigned int mss_now, gfp_t gfp) { struct tcp_sock *tp = tcp_sk(sk); @@ -1337,7 +1352,7 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, /* Link BUFF into the send queue. */ __skb_header_release(buff); - tcp_insert_write_queue_after(skb, buff, sk); + tcp_insert_write_queue_after(skb, buff, sk, tcp_queue); list_add(&buff->tcp_tsorted_anchor, &skb->tcp_tsorted_anchor); return 0; @@ -1625,10 +1640,10 @@ static void tcp_cwnd_validate(struct sock *sk, bool is_cwnd_limited) * is caused by insufficient sender buffer: * 1) just sent some data (see tcp_write_xmit) * 2) not cwnd limited (this else condition) - * 3) no more data to send (null tcp_send_head ) + * 3) no more data to send (tcp_write_queue_empty()) * 4) application is hitting buffer limit (SOCK_NOSPACE) */ - if (!tcp_send_head(sk) && sk->sk_socket && + if (tcp_write_queue_empty(sk) && sk->sk_socket && test_bit(SOCK_NOSPACE, &sk->sk_socket->flags) && (1 << sk->sk_state) & (TCPF_ESTABLISHED | TCPF_CLOSE_WAIT)) tcp_chrono_start(sk, TCP_CHRONO_SNDBUF_LIMITED); @@ -1824,7 +1839,8 @@ static bool tcp_snd_wnd_test(const struct tcp_sock *tp, * know that all the data is in scatter-gather pages, and that the * packet has never been sent out before (and thus is not cloned). */ -static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len, +static int tso_fragment(struct sock *sk, enum tcp_queue tcp_queue, + struct sk_buff *skb, unsigned int len, unsigned int mss_now, gfp_t gfp) { struct sk_buff *buff; @@ -1833,7 +1849,7 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len, /* All of a TSO frame must be composed of paged data. */ if (skb->len != skb->data_len) - return tcp_fragment(sk, skb, len, mss_now, gfp); + return tcp_fragment(sk, tcp_queue, skb, len, mss_now, gfp); buff = sk_stream_alloc_skb(sk, 0, gfp, true); if (unlikely(!buff)) @@ -1869,7 +1885,7 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len, /* Link BUFF into the send queue. */ __skb_header_release(buff); - tcp_insert_write_queue_after(skb, buff, sk); + tcp_insert_write_queue_after(skb, buff, sk, tcp_queue); return 0; } @@ -1939,8 +1955,10 @@ static bool tcp_tso_should_defer(struct sock *sk, struct sk_buff *skb, goto send_now; } - head = tcp_write_queue_head(sk); - + /* TODO : use tsorted_sent_queue ? */ + head = tcp_rtx_queue_head(sk); + if (!head) + goto send_now; age = tcp_stamp_us_delta(tp->tcp_mstamp, head->skb_mstamp); /* If next ACK is likely to come too late (half srtt), do not defer */ if (age < (tp->srtt_us >> 4)) @@ -2158,13 +2176,12 @@ static bool tcp_small_queue_check(struct sock *sk, const struct sk_buff *skb, limit <<= factor; if (refcount_read(&sk->sk_wmem_alloc) > limit) { - /* Always send the 1st or 2nd skb in write queue. + /* Always send skb if rtx queue is empty. * No need to wait for TX completion to call us back, * after softirq/tasklet schedule. * This helps when TX completions are delayed too much. */ - if (skb == sk->sk_write_queue.next || - skb->prev == sk->sk_write_queue.next) + if (tcp_rtx_queue_empty(sk)) return false; set_bit(TSQ_THROTTLED, &sk->sk_tsq_flags); @@ -2215,7 +2232,7 @@ void tcp_chrono_stop(struct sock *sk, const enum tcp_chrono type) * it's the "most interesting" or current chrono we are * tracking and starts busy chrono if we have pending data. */ - if (tcp_write_queue_empty(sk)) + if (tcp_rtx_and_write_queues_empty(sk)) tcp_chrono_set(tp, TCP_CHRONO_UNSPEC); else if (type == tp->chrono_type) tcp_chrono_set(tp, TCP_CHRONO_BUSY); @@ -2310,7 +2327,8 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, nonagle); if (skb->len > limit && - unlikely(tso_fragment(sk, skb, limit, mss_now, gfp))) + unlikely(tso_fragment(sk, TCP_FRAG_IN_WRITE_QUEUE, + skb, limit, mss_now, gfp))) break; if (test_bit(TCP_TSQ_DEFERRED, &sk->sk_tsq_flags)) @@ -2350,7 +2368,7 @@ repair: tcp_cwnd_validate(sk, is_cwnd_limited); return false; } - return !tp->packets_out && tcp_send_head(sk); + return !tp->packets_out && !tcp_write_queue_empty(sk); } bool tcp_schedule_loss_probe(struct sock *sk) @@ -2374,7 +2392,7 @@ bool tcp_schedule_loss_probe(struct sock *sk) return false; if ((tp->snd_cwnd > tcp_packets_in_flight(tp)) && - tcp_send_head(sk)) + !tcp_write_queue_empty(sk)) return false; /* Probe timeout is 2*rtt. Add minimum RTO to account @@ -2427,18 +2445,14 @@ void tcp_send_loss_probe(struct sock *sk) int mss = tcp_current_mss(sk); skb = tcp_send_head(sk); - if (skb) { - if (tcp_snd_wnd_test(tp, skb, mss)) { - pcount = tp->packets_out; - tcp_write_xmit(sk, mss, TCP_NAGLE_OFF, 2, GFP_ATOMIC); - if (tp->packets_out > pcount) - goto probe_sent; - goto rearm_timer; - } - skb = tcp_write_queue_prev(sk, skb); - } else { - skb = tcp_write_queue_tail(sk); + if (skb && tcp_snd_wnd_test(tp, skb, mss)) { + pcount = tp->packets_out; + tcp_write_xmit(sk, mss, TCP_NAGLE_OFF, 2, GFP_ATOMIC); + if (tp->packets_out > pcount) + goto probe_sent; + goto rearm_timer; } + skb = skb_rb_last(&sk->tcp_rtx_queue); /* At most one outstanding TLP retransmission. */ if (tp->tlp_high_seq) @@ -2456,10 +2470,11 @@ void tcp_send_loss_probe(struct sock *sk) goto rearm_timer; if ((pcount > 1) && (skb->len > (pcount - 1) * mss)) { - if (unlikely(tcp_fragment(sk, skb, (pcount - 1) * mss, mss, + if (unlikely(tcp_fragment(sk, TCP_FRAG_IN_RTX_QUEUE, skb, + (pcount - 1) * mss, mss, GFP_ATOMIC))) goto rearm_timer; - skb = tcp_write_queue_next(sk, skb); + skb = skb_rb_next(skb); } if (WARN_ON(!skb || !tcp_skb_pcount(skb))) @@ -2659,7 +2674,7 @@ void tcp_skb_collapse_tstamp(struct sk_buff *skb, static bool tcp_collapse_retrans(struct sock *sk, struct sk_buff *skb) { struct tcp_sock *tp = tcp_sk(sk); - struct sk_buff *next_skb = tcp_write_queue_next(sk, skb); + struct sk_buff *next_skb = skb_rb_next(skb); int skb_size, next_skb_size; skb_size = skb->len; @@ -2676,8 +2691,6 @@ static bool tcp_collapse_retrans(struct sock *sk, struct sk_buff *skb) } tcp_highest_sack_combine(sk, next_skb, skb); - tcp_unlink_write_queue(next_skb, sk); - if (next_skb->ip_summed == CHECKSUM_PARTIAL) skb->ip_summed = CHECKSUM_PARTIAL; @@ -2705,7 +2718,7 @@ static bool tcp_collapse_retrans(struct sock *sk, struct sk_buff *skb) tcp_skb_collapse_tstamp(skb, next_skb); - sk_wmem_free_skb(sk, next_skb); + tcp_rtx_queue_unlink_and_free(next_skb, sk); return true; } @@ -2716,8 +2729,6 @@ static bool tcp_can_collapse(const struct sock *sk, const struct sk_buff *skb) return false; if (skb_cloned(skb)) return false; - if (skb == tcp_send_head(sk)) - return false; /* Some heuristics for collapsing over SACK'd could be invented */ if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED) return false; @@ -2740,7 +2751,7 @@ static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *to, if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_SYN) return; - tcp_for_write_queue_from_safe(skb, tmp, sk) { + skb_rbtree_walk_from_safe(skb, tmp) { if (!tcp_can_collapse(sk, skb)) break; @@ -2815,7 +2826,8 @@ int __tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb, int segs) len = cur_mss * segs; if (skb->len > len) { - if (tcp_fragment(sk, skb, len, cur_mss, GFP_ATOMIC)) + if (tcp_fragment(sk, TCP_FRAG_IN_RTX_QUEUE, skb, len, + cur_mss, GFP_ATOMIC)) return -ENOMEM; /* We'll try again later. */ } else { if (skb_unclone(skb, GFP_ATOMIC)) @@ -2906,29 +2918,24 @@ int tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb, int segs) void tcp_xmit_retransmit_queue(struct sock *sk) { const struct inet_connection_sock *icsk = inet_csk(sk); + struct sk_buff *skb, *rtx_head = NULL, *hole = NULL; struct tcp_sock *tp = tcp_sk(sk); - struct sk_buff *skb; - struct sk_buff *hole = NULL; u32 max_segs; int mib_idx; if (!tp->packets_out) return; - if (tp->retransmit_skb_hint) { - skb = tp->retransmit_skb_hint; - } else { - skb = tcp_write_queue_head(sk); + skb = tp->retransmit_skb_hint; + if (!skb) { + rtx_head = tcp_rtx_queue_head(sk); + skb = rtx_head; } - max_segs = tcp_tso_segs(sk, tcp_current_mss(sk)); - tcp_for_write_queue_from(skb, sk) { + skb_rbtree_walk_from(skb) { __u8 sacked; int segs; - if (skb == tcp_send_head(sk)) - break; - if (tcp_pacing_check(sk)) break; @@ -2973,7 +2980,7 @@ void tcp_xmit_retransmit_queue(struct sock *sk) if (tcp_in_cwnd_reduction(sk)) tp->prr_out += tcp_skb_pcount(skb); - if (skb == tcp_write_queue_head(sk) && + if (skb == rtx_head && icsk->icsk_pending != ICSK_TIME_REO_TIMEOUT) inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, inet_csk(sk)->icsk_rto, @@ -3015,12 +3022,15 @@ void tcp_send_fin(struct sock *sk) * Note: in the latter case, FIN packet will be sent after a timeout, * as TCP stack thinks it has already been transmitted. */ - if (tskb && (tcp_send_head(sk) || tcp_under_memory_pressure(sk))) { + if (!tskb && tcp_under_memory_pressure(sk)) + tskb = skb_rb_last(&sk->tcp_rtx_queue); + + if (tskb) { coalesce: TCP_SKB_CB(tskb)->tcp_flags |= TCPHDR_FIN; TCP_SKB_CB(tskb)->end_seq++; tp->write_seq++; - if (!tcp_send_head(sk)) { + if (tcp_write_queue_empty(sk)) { /* This means tskb was already sent. * Pretend we included the FIN on previous transmit. * We need to set tp->snd_nxt to the value it would have @@ -3086,9 +3096,9 @@ int tcp_send_synack(struct sock *sk) { struct sk_buff *skb; - skb = tcp_write_queue_head(sk); + skb = tcp_rtx_queue_head(sk); if (!skb || !(TCP_SKB_CB(skb)->tcp_flags & TCPHDR_SYN)) { - pr_debug("%s: wrong queue state\n", __func__); + pr_err("%s: wrong queue state\n", __func__); return -EFAULT; } if (!(TCP_SKB_CB(skb)->tcp_flags & TCPHDR_ACK)) { @@ -3101,10 +3111,9 @@ int tcp_send_synack(struct sock *sk) if (!nskb) return -ENOMEM; INIT_LIST_HEAD(&nskb->tcp_tsorted_anchor); - tcp_unlink_write_queue(skb, sk); + tcp_rtx_queue_unlink_and_free(skb, sk); __skb_header_release(nskb); - __tcp_add_write_queue_head(sk, nskb); - sk_wmem_free_skb(sk, skb); + tcp_rbtree_insert(&sk->tcp_rtx_queue, nskb); sk->sk_wmem_queued += nskb->truesize; sk_mem_charge(sk, nskb->truesize); skb = nskb; @@ -3327,7 +3336,6 @@ static void tcp_connect_queue_skb(struct sock *sk, struct sk_buff *skb) tcb->end_seq += skb->len; __skb_header_release(skb); - __tcp_add_write_queue_tail(sk, skb); sk->sk_wmem_queued += skb->truesize; sk_mem_charge(sk, skb->truesize); tp->write_seq = tcb->end_seq; @@ -3405,12 +3413,13 @@ static int tcp_send_syn_data(struct sock *sk, struct sk_buff *syn) TCP_SKB_CB(syn_data)->tcp_flags = TCPHDR_ACK | TCPHDR_PSH; if (!err) { tp->syn_data = (fo->copied > 0); + tcp_rbtree_insert(&sk->tcp_rtx_queue, syn_data); NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPORIGDATASENT); goto done; } - /* data was not sent, this is our new send_head */ - sk->sk_send_head = syn_data; + /* data was not sent, put it in write_queue */ + __skb_queue_tail(&sk->sk_write_queue, syn_data); tp->packets_out -= tcp_skb_pcount(syn_data); fallback: @@ -3453,6 +3462,7 @@ int tcp_connect(struct sock *sk) tp->retrans_stamp = tcp_time_stamp(tp); tcp_connect_queue_skb(sk, buff); tcp_ecn_send_syn(sk, buff); + tcp_rbtree_insert(&sk->tcp_rtx_queue, buff); /* Send off SYN; include data in Fast Open. */ err = tp->fastopen_req ? tcp_send_syn_data(sk, buff) : @@ -3647,7 +3657,8 @@ int tcp_write_wakeup(struct sock *sk, int mib) skb->len > mss) { seg_size = min(seg_size, mss); TCP_SKB_CB(skb)->tcp_flags |= TCPHDR_PSH; - if (tcp_fragment(sk, skb, seg_size, mss, GFP_ATOMIC)) + if (tcp_fragment(sk, TCP_FRAG_IN_WRITE_QUEUE, + skb, seg_size, mss, GFP_ATOMIC)) return -1; } else if (!tcp_skb_pcount(skb)) tcp_set_skb_tso_segs(skb, mss); @@ -3677,7 +3688,7 @@ void tcp_send_probe0(struct sock *sk) err = tcp_write_wakeup(sk, LINUX_MIB_TCPWINPROBE); - if (tp->packets_out || !tcp_send_head(sk)) { + if (tp->packets_out || tcp_write_queue_empty(sk)) { /* Cancel probe timer, if it is not required. */ icsk->icsk_probes_out = 0; icsk->icsk_backoff = 0; diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 655dd8d7f064..7014cc00c74c 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -156,8 +156,13 @@ static bool retransmits_timed_out(struct sock *sk, return false; start_ts = tcp_sk(sk)->retrans_stamp; - if (unlikely(!start_ts)) - start_ts = tcp_skb_timestamp(tcp_write_queue_head(sk)); + if (unlikely(!start_ts)) { + struct sk_buff *head = tcp_rtx_queue_head(sk); + + if (!head) + return false; + start_ts = tcp_skb_timestamp(head); + } if (likely(timeout == 0)) { linear_backoff_thresh = ilog2(TCP_RTO_MAX/rto_base); @@ -304,11 +309,12 @@ static void tcp_delack_timer(unsigned long data) static void tcp_probe_timer(struct sock *sk) { struct inet_connection_sock *icsk = inet_csk(sk); + struct sk_buff *skb = tcp_send_head(sk); struct tcp_sock *tp = tcp_sk(sk); int max_probes; u32 start_ts; - if (tp->packets_out || !tcp_send_head(sk)) { + if (tp->packets_out || !skb) { icsk->icsk_probes_out = 0; return; } @@ -321,9 +327,9 @@ static void tcp_probe_timer(struct sock *sk) * corresponding system limit. We also implement similar policy when * we use RTO to probe window in tcp_retransmit_timer(). */ - start_ts = tcp_skb_timestamp(tcp_send_head(sk)); + start_ts = tcp_skb_timestamp(skb); if (!start_ts) - tcp_send_head(sk)->skb_mstamp = tp->tcp_mstamp; + skb->skb_mstamp = tp->tcp_mstamp; else if (icsk->icsk_user_timeout && (s32)(tcp_time_stamp(tp) - start_ts) > jiffies_to_msecs(icsk->icsk_user_timeout)) @@ -408,7 +414,7 @@ void tcp_retransmit_timer(struct sock *sk) if (!tp->packets_out) goto out; - WARN_ON(tcp_write_queue_empty(sk)); + WARN_ON(tcp_rtx_queue_empty(sk)); tp->tlp_high_seq = 0; @@ -441,7 +447,7 @@ void tcp_retransmit_timer(struct sock *sk) goto out; } tcp_enter_loss(sk); - tcp_retransmit_skb(sk, tcp_write_queue_head(sk), 1); + tcp_retransmit_skb(sk, tcp_rtx_queue_head(sk), 1); __sk_dst_reset(sk); goto out_reset_timer; } @@ -473,7 +479,7 @@ void tcp_retransmit_timer(struct sock *sk) tcp_enter_loss(sk); - if (tcp_retransmit_skb(sk, tcp_write_queue_head(sk), 1) > 0) { + if (tcp_retransmit_skb(sk, tcp_rtx_queue_head(sk), 1) > 0) { /* Retransmission failed because of local congestion, * do not backoff. */ @@ -647,7 +653,7 @@ static void tcp_keepalive_timer (unsigned long data) elapsed = keepalive_time_when(tp); /* It is alive without keepalive 8) */ - if (tp->packets_out || tcp_send_head(sk)) + if (tp->packets_out || !tcp_write_queue_empty(sk)) goto resched; elapsed = keepalive_time_elapsed(tp); -- cgit v1.2.3 From 180ca444b985c42948fa26abd278e616b5ce7eb2 Mon Sep 17 00:00:00 2001 From: Wei Wang Date: Fri, 6 Oct 2017 12:05:56 -0700 Subject: ipv6: introduce a new function fib6_update_sernum() This function takes a route as input and tries to update the sernum in the fib6_node this route is associated with. It will be used in later commit when adding a cached route into the exception table under that route. Signed-off-by: Wei Wang Signed-off-by: Martin KaFai Lau Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 2 ++ net/ipv6/ip6_fib.c | 14 ++++++++++++++ 2 files changed, 16 insertions(+) (limited to 'include/net') diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index d060d711a624..152b7b14a5a5 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -358,6 +358,8 @@ void __net_exit fib6_notifier_exit(struct net *net); unsigned int fib6_tables_seq_read(struct net *net); int fib6_tables_dump(struct net *net, struct notifier_block *nb); +void fib6_update_sernum(struct rt6_info *rt); + #ifdef CONFIG_IPV6_MULTIPLE_TABLES int fib6_rules_init(void); void fib6_rules_cleanup(void); diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index e5308d7cbd75..0ba4fbb2f855 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -110,6 +110,20 @@ enum { FIB6_NO_SERNUM_CHANGE = 0, }; +void fib6_update_sernum(struct rt6_info *rt) +{ + struct fib6_table *table = rt->rt6i_table; + struct net *net = dev_net(rt->dst.dev); + struct fib6_node *fn; + + write_lock_bh(&table->tb6_lock); + fn = rcu_dereference_protected(rt->rt6i_node, + lockdep_is_held(&table->tb6_lock)); + if (fn) + fn->fn_sernum = fib6_new_sernum(net); + write_unlock_bh(&table->tb6_lock); +} + /* * Auxiliary address test functions for the radix tree. * -- cgit v1.2.3 From 35732d01fe311ec13c4e42936878b782b8e7ea85 Mon Sep 17 00:00:00 2001 From: Wei Wang Date: Fri, 6 Oct 2017 12:05:57 -0700 Subject: ipv6: introduce a hash table to store dst cache Add a hash table into struct rt6_info in order to store dst caches created by pmtu discovery and ip redirect in ipv6 routing code. APIs to add dst cache, delete dst cache, find dst cache and update dst cache in the hash table are implemented and will be used in later commits. This is a preparation work to move all cache routes into the exception table instead of getting inserted into the fib6 tree. Signed-off-by: Wei Wang Signed-off-by: Martin KaFai Lau Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 19 +++ include/net/ip6_route.h | 3 + net/ipv6/route.c | 341 ++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 363 insertions(+) (limited to 'include/net') diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 152b7b14a5a5..c4864c1e8f13 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -98,6 +98,22 @@ struct rt6key { struct fib6_table; +struct rt6_exception_bucket { + struct hlist_head chain; + int depth; +}; + +struct rt6_exception { + struct hlist_node hlist; + struct rt6_info *rt6i; + unsigned long stamp; + struct rcu_head rcu; +}; + +#define FIB6_EXCEPTION_BUCKET_SIZE_SHIFT 10 +#define FIB6_EXCEPTION_BUCKET_SIZE (1 << FIB6_EXCEPTION_BUCKET_SIZE_SHIFT) +#define FIB6_MAX_DEPTH 5 + struct rt6_info { struct dst_entry dst; @@ -134,12 +150,15 @@ struct rt6_info { struct inet6_dev *rt6i_idev; struct rt6_info * __percpu *rt6i_pcpu; + struct rt6_exception_bucket __rcu *rt6i_exception_bucket; u32 rt6i_metric; u32 rt6i_pmtu; /* more non-fragment space at head required */ unsigned short rt6i_nfheader_len; u8 rt6i_protocol; + u8 exception_bucket_flushed:1, + unused:7; }; static inline struct inet6_dev *ip6_dst_idev(struct dst_entry *dst) diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index ee96f402cb75..3315605f34c9 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -95,6 +95,9 @@ int ip6_route_add(struct fib6_config *cfg, struct netlink_ext_ack *extack); int ip6_ins_rt(struct rt6_info *); int ip6_del_rt(struct rt6_info *); +void rt6_flush_exceptions(struct rt6_info *rt); +int rt6_remove_exception_rt(struct rt6_info *rt); + static inline int ip6_route_get_saddr(struct net *net, struct rt6_info *rt, const struct in6_addr *daddr, unsigned int prefs, diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 26cc9f483b6d..dc5e70975966 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -44,6 +44,7 @@ #include #include #include +#include #include #include #include @@ -104,6 +105,9 @@ static int rt6_fill_node(struct net *net, struct in6_addr *dst, struct in6_addr *src, int iif, int type, u32 portid, u32 seq, unsigned int flags); +static struct rt6_info *rt6_find_cached_rt(struct rt6_info *rt, + struct in6_addr *daddr, + struct in6_addr *saddr); #ifdef CONFIG_IPV6_ROUTE_INFO static struct rt6_info *rt6_add_route_info(struct net *net, @@ -392,6 +396,7 @@ EXPORT_SYMBOL(ip6_dst_alloc); static void ip6_dst_destroy(struct dst_entry *dst) { struct rt6_info *rt = (struct rt6_info *)dst; + struct rt6_exception_bucket *bucket; struct dst_entry *from = dst->from; struct inet6_dev *idev; @@ -404,6 +409,11 @@ static void ip6_dst_destroy(struct dst_entry *dst) rt->rt6i_idev = NULL; in6_dev_put(idev); } + bucket = rcu_dereference_protected(rt->rt6i_exception_bucket, 1); + if (bucket) { + rt->rt6i_exception_bucket = NULL; + kfree(bucket); + } dst->from = NULL; dst_release(from); @@ -1091,6 +1101,337 @@ static struct rt6_info *rt6_make_pcpu_route(struct rt6_info *rt) return pcpu_rt; } +/* exception hash table implementation + */ +static DEFINE_SPINLOCK(rt6_exception_lock); + +/* Remove rt6_ex from hash table and free the memory + * Caller must hold rt6_exception_lock + */ +static void rt6_remove_exception(struct rt6_exception_bucket *bucket, + struct rt6_exception *rt6_ex) +{ + if (!bucket || !rt6_ex) + return; + rt6_ex->rt6i->rt6i_node = NULL; + hlist_del_rcu(&rt6_ex->hlist); + rt6_release(rt6_ex->rt6i); + kfree_rcu(rt6_ex, rcu); + WARN_ON_ONCE(!bucket->depth); + bucket->depth--; +} + +/* Remove oldest rt6_ex in bucket and free the memory + * Caller must hold rt6_exception_lock + */ +static void rt6_exception_remove_oldest(struct rt6_exception_bucket *bucket) +{ + struct rt6_exception *rt6_ex, *oldest = NULL; + + if (!bucket) + return; + + hlist_for_each_entry(rt6_ex, &bucket->chain, hlist) { + if (!oldest || time_before(rt6_ex->stamp, oldest->stamp)) + oldest = rt6_ex; + } + rt6_remove_exception(bucket, oldest); +} + +static u32 rt6_exception_hash(const struct in6_addr *dst, + const struct in6_addr *src) +{ + static u32 seed __read_mostly; + u32 val; + + net_get_random_once(&seed, sizeof(seed)); + val = jhash(dst, sizeof(*dst), seed); + +#ifdef CONFIG_IPV6_SUBTREES + if (src) + val = jhash(src, sizeof(*src), val); +#endif + return hash_32(val, FIB6_EXCEPTION_BUCKET_SIZE_SHIFT); +} + +/* Helper function to find the cached rt in the hash table + * and update bucket pointer to point to the bucket for this + * (daddr, saddr) pair + * Caller must hold rt6_exception_lock + */ +static struct rt6_exception * +__rt6_find_exception_spinlock(struct rt6_exception_bucket **bucket, + const struct in6_addr *daddr, + const struct in6_addr *saddr) +{ + struct rt6_exception *rt6_ex; + u32 hval; + + if (!(*bucket) || !daddr) + return NULL; + + hval = rt6_exception_hash(daddr, saddr); + *bucket += hval; + + hlist_for_each_entry(rt6_ex, &(*bucket)->chain, hlist) { + struct rt6_info *rt6 = rt6_ex->rt6i; + bool matched = ipv6_addr_equal(daddr, &rt6->rt6i_dst.addr); + +#ifdef CONFIG_IPV6_SUBTREES + if (matched && saddr) + matched = ipv6_addr_equal(saddr, &rt6->rt6i_src.addr); +#endif + if (matched) + return rt6_ex; + } + return NULL; +} + +/* Helper function to find the cached rt in the hash table + * and update bucket pointer to point to the bucket for this + * (daddr, saddr) pair + * Caller must hold rcu_read_lock() + */ +static struct rt6_exception * +__rt6_find_exception_rcu(struct rt6_exception_bucket **bucket, + const struct in6_addr *daddr, + const struct in6_addr *saddr) +{ + struct rt6_exception *rt6_ex; + u32 hval; + + WARN_ON_ONCE(!rcu_read_lock_held()); + + if (!(*bucket) || !daddr) + return NULL; + + hval = rt6_exception_hash(daddr, saddr); + *bucket += hval; + + hlist_for_each_entry_rcu(rt6_ex, &(*bucket)->chain, hlist) { + struct rt6_info *rt6 = rt6_ex->rt6i; + bool matched = ipv6_addr_equal(daddr, &rt6->rt6i_dst.addr); + +#ifdef CONFIG_IPV6_SUBTREES + if (matched && saddr) + matched = ipv6_addr_equal(saddr, &rt6->rt6i_src.addr); +#endif + if (matched) + return rt6_ex; + } + return NULL; +} + +static int rt6_insert_exception(struct rt6_info *nrt, + struct rt6_info *ort) +{ + struct rt6_exception_bucket *bucket; + struct in6_addr *src_key = NULL; + struct rt6_exception *rt6_ex; + int err = 0; + + /* ort can't be a cache or pcpu route */ + if (ort->rt6i_flags & (RTF_CACHE | RTF_PCPU)) + ort = (struct rt6_info *)ort->dst.from; + WARN_ON_ONCE(ort->rt6i_flags & (RTF_CACHE | RTF_PCPU)); + + spin_lock_bh(&rt6_exception_lock); + + if (ort->exception_bucket_flushed) { + err = -EINVAL; + goto out; + } + + bucket = rcu_dereference_protected(ort->rt6i_exception_bucket, + lockdep_is_held(&rt6_exception_lock)); + if (!bucket) { + bucket = kcalloc(FIB6_EXCEPTION_BUCKET_SIZE, sizeof(*bucket), + GFP_ATOMIC); + if (!bucket) { + err = -ENOMEM; + goto out; + } + rcu_assign_pointer(ort->rt6i_exception_bucket, bucket); + } + +#ifdef CONFIG_IPV6_SUBTREES + /* rt6i_src.plen != 0 indicates ort is in subtree + * and exception table is indexed by a hash of + * both rt6i_dst and rt6i_src. + * Otherwise, the exception table is indexed by + * a hash of only rt6i_dst. + */ + if (ort->rt6i_src.plen) + src_key = &nrt->rt6i_src.addr; +#endif + rt6_ex = __rt6_find_exception_spinlock(&bucket, &nrt->rt6i_dst.addr, + src_key); + if (rt6_ex) + rt6_remove_exception(bucket, rt6_ex); + + rt6_ex = kzalloc(sizeof(*rt6_ex), GFP_ATOMIC); + if (!rt6_ex) { + err = -ENOMEM; + goto out; + } + rt6_ex->rt6i = nrt; + rt6_ex->stamp = jiffies; + atomic_inc(&nrt->rt6i_ref); + nrt->rt6i_node = ort->rt6i_node; + hlist_add_head_rcu(&rt6_ex->hlist, &bucket->chain); + bucket->depth++; + + if (bucket->depth > FIB6_MAX_DEPTH) + rt6_exception_remove_oldest(bucket); + +out: + spin_unlock_bh(&rt6_exception_lock); + + /* Update fn->fn_sernum to invalidate all cached dst */ + if (!err) + fib6_update_sernum(ort); + + return err; +} + +void rt6_flush_exceptions(struct rt6_info *rt) +{ + struct rt6_exception_bucket *bucket; + struct rt6_exception *rt6_ex; + struct hlist_node *tmp; + int i; + + spin_lock_bh(&rt6_exception_lock); + /* Prevent rt6_insert_exception() to recreate the bucket list */ + rt->exception_bucket_flushed = 1; + + bucket = rcu_dereference_protected(rt->rt6i_exception_bucket, + lockdep_is_held(&rt6_exception_lock)); + if (!bucket) + goto out; + + for (i = 0; i < FIB6_EXCEPTION_BUCKET_SIZE; i++) { + hlist_for_each_entry_safe(rt6_ex, tmp, &bucket->chain, hlist) + rt6_remove_exception(bucket, rt6_ex); + WARN_ON_ONCE(bucket->depth); + bucket++; + } + +out: + spin_unlock_bh(&rt6_exception_lock); +} + +/* Find cached rt in the hash table inside passed in rt + * Caller has to hold rcu_read_lock() + */ +static struct rt6_info *rt6_find_cached_rt(struct rt6_info *rt, + struct in6_addr *daddr, + struct in6_addr *saddr) +{ + struct rt6_exception_bucket *bucket; + struct in6_addr *src_key = NULL; + struct rt6_exception *rt6_ex; + struct rt6_info *res = NULL; + + bucket = rcu_dereference(rt->rt6i_exception_bucket); + +#ifdef CONFIG_IPV6_SUBTREES + /* rt6i_src.plen != 0 indicates rt is in subtree + * and exception table is indexed by a hash of + * both rt6i_dst and rt6i_src. + * Otherwise, the exception table is indexed by + * a hash of only rt6i_dst. + */ + if (rt->rt6i_src.plen) + src_key = saddr; +#endif + rt6_ex = __rt6_find_exception_rcu(&bucket, daddr, src_key); + + if (rt6_ex && !rt6_check_expired(rt6_ex->rt6i)) + res = rt6_ex->rt6i; + + return res; +} + +/* Remove the passed in cached rt from the hash table that contains it */ +int rt6_remove_exception_rt(struct rt6_info *rt) +{ + struct rt6_info *from = (struct rt6_info *)rt->dst.from; + struct rt6_exception_bucket *bucket; + struct in6_addr *src_key = NULL; + struct rt6_exception *rt6_ex; + int err; + + if (!from || + !(rt->rt6i_flags | RTF_CACHE)) + return -EINVAL; + + if (!rcu_access_pointer(from->rt6i_exception_bucket)) + return -ENOENT; + + spin_lock_bh(&rt6_exception_lock); + bucket = rcu_dereference_protected(from->rt6i_exception_bucket, + lockdep_is_held(&rt6_exception_lock)); +#ifdef CONFIG_IPV6_SUBTREES + /* rt6i_src.plen != 0 indicates 'from' is in subtree + * and exception table is indexed by a hash of + * both rt6i_dst and rt6i_src. + * Otherwise, the exception table is indexed by + * a hash of only rt6i_dst. + */ + if (from->rt6i_src.plen) + src_key = &rt->rt6i_src.addr; +#endif + rt6_ex = __rt6_find_exception_spinlock(&bucket, + &rt->rt6i_dst.addr, + src_key); + if (rt6_ex) { + rt6_remove_exception(bucket, rt6_ex); + err = 0; + } else { + err = -ENOENT; + } + + spin_unlock_bh(&rt6_exception_lock); + return err; +} + +/* Find rt6_ex which contains the passed in rt cache and + * refresh its stamp + */ +static void rt6_update_exception_stamp_rt(struct rt6_info *rt) +{ + struct rt6_info *from = (struct rt6_info *)rt->dst.from; + struct rt6_exception_bucket *bucket; + struct in6_addr *src_key = NULL; + struct rt6_exception *rt6_ex; + + if (!from || + !(rt->rt6i_flags | RTF_CACHE)) + return; + + rcu_read_lock(); + bucket = rcu_dereference(from->rt6i_exception_bucket); + +#ifdef CONFIG_IPV6_SUBTREES + /* rt6i_src.plen != 0 indicates 'from' is in subtree + * and exception table is indexed by a hash of + * both rt6i_dst and rt6i_src. + * Otherwise, the exception table is indexed by + * a hash of only rt6i_dst. + */ + if (from->rt6i_src.plen) + src_key = &rt->rt6i_src.addr; +#endif + rt6_ex = __rt6_find_exception_rcu(&bucket, + &rt->rt6i_dst.addr, + src_key); + if (rt6_ex) + rt6_ex->stamp = jiffies; + + rcu_read_unlock(); +} + struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table, int oif, struct flowi6 *fl6, int flags) { -- cgit v1.2.3 From c757faa8bfa26a0dd24b41ff783e0da042156887 Mon Sep 17 00:00:00 2001 From: Wei Wang Date: Fri, 6 Oct 2017 12:06:01 -0700 Subject: ipv6: prepare fib6_age() for exception table If all dst cache entries are stored in the exception table under the main route, we have to go through them during fib6_age() when doing garbage collecting. Introduce a new function rt6_age_exception() which goes through all dst entries in the exception table and remove those entries that are expired. This function is called in fib6_age() so that all dst caches are also garbage collected. Signed-off-by: Wei Wang Signed-off-by: Martin KaFai Lau Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 13 +++++++++++ include/net/ip6_route.h | 2 ++ net/ipv6/ip6_fib.c | 26 ++++++++------------- net/ipv6/route.c | 60 +++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 84 insertions(+), 17 deletions(-) (limited to 'include/net') diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index c4864c1e8f13..11a79ef87a28 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -29,6 +29,14 @@ #define FIB6_TABLE_HASHSZ 1 #endif +#define RT6_DEBUG 2 + +#if RT6_DEBUG >= 3 +#define RT6_TRACE(x...) pr_debug(x) +#else +#define RT6_TRACE(x...) do { ; } while (0) +#endif + struct rt6_info; struct fib6_config { @@ -75,6 +83,11 @@ struct fib6_node { struct rcu_head rcu; }; +struct fib6_gc_args { + int timeout; + int more; +}; + #ifndef CONFIG_IPV6_SUBTREES #define FIB6_SUBTREE(fn) NULL #else diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 3315605f34c9..a0087fb9864b 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -97,6 +97,8 @@ int ip6_del_rt(struct rt6_info *); void rt6_flush_exceptions(struct rt6_info *rt); int rt6_remove_exception_rt(struct rt6_info *rt); +void rt6_age_exceptions(struct rt6_info *rt, struct fib6_gc_args *gc_args, + unsigned long now); static inline int ip6_route_get_saddr(struct net *net, struct rt6_info *rt, const struct in6_addr *daddr, diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 0ba4fbb2f855..3afbe50f2779 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -38,14 +38,6 @@ #include #include -#define RT6_DEBUG 2 - -#if RT6_DEBUG >= 3 -#define RT6_TRACE(x...) pr_debug(x) -#else -#define RT6_TRACE(x...) do { ; } while (0) -#endif - static struct kmem_cache *fib6_node_kmem __read_mostly; struct fib6_cleaner { @@ -1890,12 +1882,6 @@ static void fib6_flush_trees(struct net *net) * Garbage collection */ -struct fib6_gc_args -{ - int timeout; - int more; -}; - static int fib6_age(struct rt6_info *rt, void *arg) { struct fib6_gc_args *gc_args = arg; @@ -1904,9 +1890,6 @@ static int fib6_age(struct rt6_info *rt, void *arg) /* * check addrconf expiration here. * Routes are expired even if they are in use. - * - * Also age clones. Note, that clones are aged out - * only if they are not in use now. */ if (rt->rt6i_flags & RTF_EXPIRES && rt->dst.expires) { @@ -1915,6 +1898,9 @@ static int fib6_age(struct rt6_info *rt, void *arg) return -1; } gc_args->more++; + /* The following part will soon be removed when the exception + * table is hooked up to store all cached routes. + */ } else if (rt->rt6i_flags & RTF_CACHE) { if (time_after_eq(now, rt->dst.lastuse + gc_args->timeout)) rt->dst.obsolete = DST_OBSOLETE_KILL; @@ -1940,6 +1926,12 @@ static int fib6_age(struct rt6_info *rt, void *arg) gc_args->more++; } + /* Also age clones in the exception table. + * Note, that clones are aged out + * only if they are not in use now. + */ + rt6_age_exceptions(rt, gc_args, now); + return 0; } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index e8e901589564..d2dd55f58b5d 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1528,6 +1528,66 @@ static void rt6_exceptions_clean_tohost(struct rt6_info *rt, spin_unlock_bh(&rt6_exception_lock); } +static void rt6_age_examine_exception(struct rt6_exception_bucket *bucket, + struct rt6_exception *rt6_ex, + struct fib6_gc_args *gc_args, + unsigned long now) +{ + struct rt6_info *rt = rt6_ex->rt6i; + + if (atomic_read(&rt->dst.__refcnt) == 1 && + time_after_eq(now, rt->dst.lastuse + gc_args->timeout)) { + RT6_TRACE("aging clone %p\n", rt); + rt6_remove_exception(bucket, rt6_ex); + return; + } else if (rt->rt6i_flags & RTF_GATEWAY) { + struct neighbour *neigh; + __u8 neigh_flags = 0; + + neigh = dst_neigh_lookup(&rt->dst, &rt->rt6i_gateway); + if (neigh) { + neigh_flags = neigh->flags; + neigh_release(neigh); + } + if (!(neigh_flags & NTF_ROUTER)) { + RT6_TRACE("purging route %p via non-router but gateway\n", + rt); + rt6_remove_exception(bucket, rt6_ex); + return; + } + } + gc_args->more++; +} + +void rt6_age_exceptions(struct rt6_info *rt, + struct fib6_gc_args *gc_args, + unsigned long now) +{ + struct rt6_exception_bucket *bucket; + struct rt6_exception *rt6_ex; + struct hlist_node *tmp; + int i; + + if (!rcu_access_pointer(rt->rt6i_exception_bucket)) + return; + + spin_lock_bh(&rt6_exception_lock); + bucket = rcu_dereference_protected(rt->rt6i_exception_bucket, + lockdep_is_held(&rt6_exception_lock)); + + if (bucket) { + for (i = 0; i < FIB6_EXCEPTION_BUCKET_SIZE; i++) { + hlist_for_each_entry_safe(rt6_ex, tmp, + &bucket->chain, hlist) { + rt6_age_examine_exception(bucket, rt6_ex, + gc_args, now); + } + bucket++; + } + } + spin_unlock_bh(&rt6_exception_lock); +} + struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table, int oif, struct flowi6 *fl6, int flags) { -- cgit v1.2.3 From 38fbeeeeccdb38d0635398e8e344d245f6d8dc52 Mon Sep 17 00:00:00 2001 From: Wei Wang Date: Fri, 6 Oct 2017 12:06:02 -0700 Subject: ipv6: prepare fib6_locate() for exception table fib6_locate() is used to find the fib6_node according to the passed in prefix address key. It currently tries to find the fib6_node with the exact match of the passed in key. However, when we move cached routes into the exception table, fib6_locate() will fail to find the fib6_node for it as the cached routes will be stored in the exception table under the fib6_node with the longest prefix match of the cache's dst addr key. This commit adds a new parameter to let the caller specify if it needs exact match or longest prefix match. Right now, all callers still does exact match when calling fib6_locate(). It will be changed in later commit where exception table is hooked up to store cached routes. Signed-off-by: Wei Wang Signed-off-by: Martin KaFai Lau Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 3 ++- net/ipv6/addrconf.c | 2 +- net/ipv6/ip6_fib.c | 30 +++++++++++++++++++++++------- net/ipv6/route.c | 5 +++-- 4 files changed, 29 insertions(+), 11 deletions(-) (limited to 'include/net') diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 11a79ef87a28..4497a1eb4d41 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -357,7 +357,8 @@ struct fib6_node *fib6_lookup(struct fib6_node *root, struct fib6_node *fib6_locate(struct fib6_node *root, const struct in6_addr *daddr, int dst_len, - const struct in6_addr *saddr, int src_len); + const struct in6_addr *saddr, int src_len, + bool exact_match); void fib6_clean_all(struct net *net, int (*func)(struct rt6_info *, void *arg), void *arg); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 837418ff2d4b..3ccaf52824c9 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -2322,7 +2322,7 @@ static struct rt6_info *addrconf_get_prefix_route(const struct in6_addr *pfx, return NULL; read_lock_bh(&table->tb6_lock); - fn = fib6_locate(&table->tb6_root, pfx, plen, NULL, 0); + fn = fib6_locate(&table->tb6_root, pfx, plen, NULL, 0, true); if (!fn) goto out; diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 3afbe50f2779..b3e4cf0962f8 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -1343,14 +1343,21 @@ struct fib6_node *fib6_lookup(struct fib6_node *root, const struct in6_addr *dad /* * Get node with specified destination prefix (and source prefix, * if subtrees are used) + * exact_match == true means we try to find fn with exact match of + * the passed in prefix addr + * exact_match == false means we try to find fn with longest prefix + * match of the passed in prefix addr. This is useful for finding fn + * for cached route as it will be stored in the exception table under + * the node with longest prefix length. */ static struct fib6_node *fib6_locate_1(struct fib6_node *root, const struct in6_addr *addr, - int plen, int offset) + int plen, int offset, + bool exact_match) { - struct fib6_node *fn; + struct fib6_node *fn, *prev = NULL; for (fn = root; fn ; ) { struct rt6key *key = (struct rt6key *)((u8 *)fn->leaf + offset); @@ -1360,11 +1367,13 @@ static struct fib6_node *fib6_locate_1(struct fib6_node *root, */ if (plen < fn->fn_bit || !ipv6_prefix_equal(&key->addr, addr, fn->fn_bit)) - return NULL; + goto out; if (plen == fn->fn_bit) return fn; + prev = fn; + /* * We have more bits to go */ @@ -1373,24 +1382,31 @@ static struct fib6_node *fib6_locate_1(struct fib6_node *root, else fn = fn->left; } - return NULL; +out: + if (exact_match) + return NULL; + else + return prev; } struct fib6_node *fib6_locate(struct fib6_node *root, const struct in6_addr *daddr, int dst_len, - const struct in6_addr *saddr, int src_len) + const struct in6_addr *saddr, int src_len, + bool exact_match) { struct fib6_node *fn; fn = fib6_locate_1(root, daddr, dst_len, - offsetof(struct rt6_info, rt6i_dst)); + offsetof(struct rt6_info, rt6i_dst), + exact_match); #ifdef CONFIG_IPV6_SUBTREES if (src_len) { WARN_ON(saddr == NULL); if (fn && fn->subtree) fn = fib6_locate_1(fn->subtree, saddr, src_len, - offsetof(struct rt6_info, rt6i_src)); + offsetof(struct rt6_info, rt6i_src), + exact_match); } #endif diff --git a/net/ipv6/route.c b/net/ipv6/route.c index d2dd55f58b5d..855b4ceec349 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2800,7 +2800,8 @@ static int ip6_route_del(struct fib6_config *cfg, fn = fib6_locate(&table->tb6_root, &cfg->fc_dst, cfg->fc_dst_len, - &cfg->fc_src, cfg->fc_src_len); + &cfg->fc_src, cfg->fc_src_len, + true); if (fn) { for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) { @@ -3009,7 +3010,7 @@ static struct rt6_info *rt6_get_route_info(struct net *net, return NULL; read_lock_bh(&table->tb6_lock); - fn = fib6_locate(&table->tb6_root, prefix, prefixlen, NULL, 0); + fn = fib6_locate(&table->tb6_root, prefix, prefixlen, NULL, 0, true); if (!fn) goto out; -- cgit v1.2.3 From 2b760fcf5cfb34e8610df56d83745b2b74ae1379 Mon Sep 17 00:00:00 2001 From: Wei Wang Date: Fri, 6 Oct 2017 12:06:03 -0700 Subject: ipv6: hook up exception table to store dst cache This commit makes use of the exception hash table implementation to store dst caches created by pmtu discovery and ip redirect into the hash table under the rt_info and no longer inserts these routes into fib6 tree. This makes the fib6 tree only contain static configured routes and could now be protected by rcu instead of a rw lock. With this change, in the route lookup related functions, after finding the rt6_info with the longest prefix, we also need to search for the exception table before doing backtracking. In the route delete function, if the route being deleted is not a dst cache, deletion of this route also need to flush the whole hash table under it. If it is a dst cache, then only delete the cached dst in the hash table. Note: for fib6_walk_continue() function, w->root now is always pointing to a root node considering that fib6_prune_clones() is removed from the code. So we add a WARN_ON() msg to make sure w->root always points to a root node and also removed the update of w->root in fib6_repair_tree(). This is a prerequisite for later patch because we don't need to make w->root as rcu protected when replacing rwlock with RCU. Also, we remove all prune related variables as it is no longer used. Signed-off-by: Wei Wang Signed-off-by: Martin KaFai Lau Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 1 - net/ipv6/addrconf.c | 1 - net/ipv6/ip6_fib.c | 95 ++++++++------------------------------------ net/ipv6/route.c | 108 +++++++++++++++++++++++++------------------------- 4 files changed, 72 insertions(+), 133 deletions(-) (limited to 'include/net') diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 4497a1eb4d41..d0b7283073e3 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -280,7 +280,6 @@ struct fib6_walker { struct fib6_node *root, *node; struct rt6_info *leaf; enum fib6_walk_state state; - bool prune; unsigned int skip; unsigned int count; int (*func)(struct fib6_walker *); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 3ccaf52824c9..873afafddfc4 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -2326,7 +2326,6 @@ static struct rt6_info *addrconf_get_prefix_route(const struct in6_addr *pfx, if (!fn) goto out; - noflags |= RTF_CACHE; for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) { if (rt->dst.dev->ifindex != dev->ifindex) continue; diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index b3e4cf0962f8..9c8e704e6af7 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -54,7 +54,6 @@ struct fib6_cleaner { #define FWS_INIT FWS_L #endif -static void fib6_prune_clones(struct net *net, struct fib6_node *fn); static struct rt6_info *fib6_find_prefix(struct net *net, struct fib6_node *fn); static struct fib6_node *fib6_repair_tree(struct net *net, struct fib6_node *fn); static int fib6_walk(struct net *net, struct fib6_walker *w); @@ -1101,6 +1100,8 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, if (WARN_ON_ONCE(!atomic_read(&rt->dst.__refcnt))) return -EINVAL; + if (WARN_ON_ONCE(rt->rt6i_flags & RTF_CACHE)) + return -EINVAL; if (info->nlh) { if (!(info->nlh->nlmsg_flags & NLM_F_CREATE)) @@ -1192,11 +1193,8 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, #endif err = fib6_add_rt2node(fn, rt, info, mxc); - if (!err) { + if (!err) fib6_start_gc(info->nl_net, rt); - if (!(rt->rt6i_flags & RTF_CACHE)) - fib6_prune_clones(info->nl_net, pn); - } out: if (err) { @@ -1511,19 +1509,12 @@ static struct fib6_node *fib6_repair_tree(struct net *net, read_lock(&net->ipv6.fib6_walker_lock); FOR_WALKERS(net, w) { if (!child) { - if (w->root == fn) { - w->root = w->node = NULL; - RT6_TRACE("W %p adjusted by delroot 1\n", w); - } else if (w->node == fn) { + if (w->node == fn) { RT6_TRACE("W %p adjusted by delnode 1, s=%d/%d\n", w, w->state, nstate); w->node = pn; w->state = nstate; } } else { - if (w->root == fn) { - w->root = child; - RT6_TRACE("W %p adjusted by delroot 2\n", w); - } if (w->node == fn) { w->node = child; if (children&2) { @@ -1557,12 +1548,17 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp, RT6_TRACE("fib6_del_route\n"); + WARN_ON_ONCE(rt->rt6i_flags & RTF_CACHE); + /* Unlink it */ *rtp = rt->dst.rt6_next; rt->rt6i_node = NULL; net->ipv6.rt6_stats->fib_rt_entries--; net->ipv6.rt6_stats->fib_discarded_routes++; + /* Flush all cached dst in exception table */ + rt6_flush_exceptions(rt); + /* Reset round-robin state, if necessary */ if (fn->rr_ptr == rt) fn->rr_ptr = NULL; @@ -1625,18 +1621,9 @@ int fib6_del(struct rt6_info *rt, struct nl_info *info) WARN_ON(!(fn->fn_flags & RTN_RTINFO)); - if (!(rt->rt6i_flags & RTF_CACHE)) { - struct fib6_node *pn = fn; -#ifdef CONFIG_IPV6_SUBTREES - /* clones of this route might be in another subtree */ - if (rt->rt6i_src.plen) { - while (!(pn->fn_flags & RTN_ROOT)) - pn = pn->parent; - pn = pn->parent; - } -#endif - fib6_prune_clones(info->nl_net, pn); - } + /* remove cached dst from exception table */ + if (rt->rt6i_flags & RTF_CACHE) + return rt6_remove_exception_rt(rt); /* * Walk the leaf entries looking for ourself @@ -1679,16 +1666,14 @@ static int fib6_walk_continue(struct fib6_walker *w) { struct fib6_node *fn, *pn; + /* w->root should always be table->tb6_root */ + WARN_ON_ONCE(!(w->root->fn_flags & RTN_TL_ROOT)); + for (;;) { fn = w->node; if (!fn) return 0; - if (w->prune && fn != w->root && - fn->fn_flags & RTN_RTINFO && w->state < FWS_C) { - w->state = FWS_C; - w->leaf = fn->leaf; - } switch (w->state) { #ifdef CONFIG_IPV6_SUBTREES case FWS_S: @@ -1820,20 +1805,16 @@ static int fib6_clean_node(struct fib6_walker *w) * func is called on each route. * It may return -1 -> delete this route. * 0 -> continue walking - * - * prune==1 -> only immediate children of node (certainly, - * ignoring pure split nodes) will be scanned. */ static void fib6_clean_tree(struct net *net, struct fib6_node *root, int (*func)(struct rt6_info *, void *arg), - bool prune, int sernum, void *arg) + int sernum, void *arg) { struct fib6_cleaner c; c.w.root = root; c.w.func = fib6_clean_node; - c.w.prune = prune; c.w.count = 0; c.w.skip = 0; c.func = func; @@ -1858,7 +1839,7 @@ static void __fib6_clean_all(struct net *net, hlist_for_each_entry_rcu(table, head, tb6_hlist) { write_lock_bh(&table->tb6_lock); fib6_clean_tree(net, &table->tb6_root, - func, false, sernum, arg); + func, sernum, arg); write_unlock_bh(&table->tb6_lock); } } @@ -1871,22 +1852,6 @@ void fib6_clean_all(struct net *net, int (*func)(struct rt6_info *, void *), __fib6_clean_all(net, func, FIB6_NO_SERNUM_CHANGE, arg); } -static int fib6_prune_clone(struct rt6_info *rt, void *arg) -{ - if (rt->rt6i_flags & RTF_CACHE) { - RT6_TRACE("pruning clone %p\n", rt); - return -1; - } - - return 0; -} - -static void fib6_prune_clones(struct net *net, struct fib6_node *fn) -{ - fib6_clean_tree(net, fn, fib6_prune_clone, true, - FIB6_NO_SERNUM_CHANGE, NULL); -} - static void fib6_flush_trees(struct net *net) { int new_sernum = fib6_new_sernum(net); @@ -1914,32 +1879,6 @@ static int fib6_age(struct rt6_info *rt, void *arg) return -1; } gc_args->more++; - /* The following part will soon be removed when the exception - * table is hooked up to store all cached routes. - */ - } else if (rt->rt6i_flags & RTF_CACHE) { - if (time_after_eq(now, rt->dst.lastuse + gc_args->timeout)) - rt->dst.obsolete = DST_OBSOLETE_KILL; - if (atomic_read(&rt->dst.__refcnt) == 1 && - rt->dst.obsolete == DST_OBSOLETE_KILL) { - RT6_TRACE("aging clone %p\n", rt); - return -1; - } else if (rt->rt6i_flags & RTF_GATEWAY) { - struct neighbour *neigh; - __u8 neigh_flags = 0; - - neigh = dst_neigh_lookup(&rt->dst, &rt->rt6i_gateway); - if (neigh) { - neigh_flags = neigh->flags; - neigh_release(neigh); - } - if (!(neigh_flags & NTF_ROUTER)) { - RT6_TRACE("purging route %p via non-router but gateway\n", - rt); - return -1; - } - } - gc_args->more++; } /* Also age clones in the exception table. diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 855b4ceec349..65130dde276a 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -878,8 +878,8 @@ static struct rt6_info *ip6_pol_route_lookup(struct net *net, struct fib6_table *table, struct flowi6 *fl6, int flags) { + struct rt6_info *rt, *rt_cache; struct fib6_node *fn; - struct rt6_info *rt; read_lock_bh(&table->tb6_lock); fn = fib6_lookup(&table->tb6_root, &fl6->daddr, &fl6->saddr); @@ -893,6 +893,11 @@ restart: if (fn) goto restart; } + /* Search through exception table */ + rt_cache = rt6_find_cached_rt(rt, &fl6->daddr, &fl6->saddr); + if (rt_cache) + rt = rt_cache; + dst_use(&rt->dst, jiffies); read_unlock_bh(&table->tb6_lock); @@ -1592,7 +1597,7 @@ struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table, int oif, struct flowi6 *fl6, int flags) { struct fib6_node *fn, *saved_fn; - struct rt6_info *rt; + struct rt6_info *rt, *rt_cache; int strict = 0; strict |= flags & RT6_LOOKUP_F_IFACE; @@ -1624,6 +1629,10 @@ redo_rt6_select: } } + /*Search through exception table */ + rt_cache = rt6_find_cached_rt(rt, &fl6->daddr, &fl6->saddr); + if (rt_cache) + rt = rt_cache; if (rt == net->ipv6.ip6_null_entry || (rt->rt6i_flags & RTF_CACHE)) { dst_use(&rt->dst, jiffies); @@ -1988,23 +1997,17 @@ static void __ip6_rt_update_pmtu(struct dst_entry *dst, const struct sock *sk, if (!rt6_cache_allowed_for_pmtu(rt6)) { rt6_do_update_pmtu(rt6, mtu); + /* update rt6_ex->stamp for cache */ + if (rt6->rt6i_flags & RTF_CACHE) + rt6_update_exception_stamp_rt(rt6); } else if (daddr) { struct rt6_info *nrt6; nrt6 = ip6_rt_cache_alloc(rt6, daddr, saddr); if (nrt6) { rt6_do_update_pmtu(nrt6, mtu); - - /* ip6_ins_rt(nrt6) will bump the - * rt6->rt6i_node->fn_sernum - * which will fail the next rt6_check() and - * invalidate the sk->sk_dst_cache. - */ - ip6_ins_rt(nrt6); - /* Release the reference taken in - * ip6_rt_cache_alloc() - */ - dst_release(&nrt6->dst); + if (rt6_insert_exception(nrt6, rt6)) + dst_release_immediate(&nrt6->dst); } } } @@ -2068,7 +2071,7 @@ static struct rt6_info *__ip6_route_redirect(struct net *net, int flags) { struct ip6rd_flowi *rdfl = (struct ip6rd_flowi *)fl6; - struct rt6_info *rt; + struct rt6_info *rt, *rt_cache; struct fib6_node *fn; /* Get the "current" route for this destination and @@ -2093,8 +2096,23 @@ restart: continue; if (fl6->flowi6_oif != rt->dst.dev->ifindex) continue; - if (!ipv6_addr_equal(&rdfl->gateway, &rt->rt6i_gateway)) + /* rt_cache's gateway might be different from its 'parent' + * in the case of an ip redirect. + * So we keep searching in the exception table if the gateway + * is different. + */ + if (!ipv6_addr_equal(&rdfl->gateway, &rt->rt6i_gateway)) { + rt_cache = rt6_find_cached_rt(rt, + &fl6->daddr, + &fl6->saddr); + if (rt_cache && + ipv6_addr_equal(&rdfl->gateway, + &rt_cache->rt6i_gateway)) { + rt = rt_cache; + break; + } continue; + } break; } @@ -2785,9 +2803,9 @@ out_put: static int ip6_route_del(struct fib6_config *cfg, struct netlink_ext_ack *extack) { + struct rt6_info *rt, *rt_cache; struct fib6_table *table; struct fib6_node *fn; - struct rt6_info *rt; int err = -ESRCH; table = fib6_get_table(cfg->fc_nlinfo.nl_net, cfg->fc_table); @@ -2801,13 +2819,17 @@ static int ip6_route_del(struct fib6_config *cfg, fn = fib6_locate(&table->tb6_root, &cfg->fc_dst, cfg->fc_dst_len, &cfg->fc_src, cfg->fc_src_len, - true); + !(cfg->fc_flags & RTF_CACHE)); if (fn) { for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) { - if ((rt->rt6i_flags & RTF_CACHE) && - !(cfg->fc_flags & RTF_CACHE)) - continue; + if (cfg->fc_flags & RTF_CACHE) { + rt_cache = rt6_find_cached_rt(rt, &cfg->fc_dst, + &cfg->fc_src); + if (!rt_cache) + continue; + rt = rt_cache; + } if (cfg->fc_ifindex && (!rt->dst.dev || rt->dst.dev->ifindex != cfg->fc_ifindex)) @@ -2933,8 +2955,14 @@ static void rt6_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_bu nrt->rt6i_protocol = RTPROT_REDIRECT; nrt->rt6i_gateway = *(struct in6_addr *)neigh->primary_key; - if (ip6_ins_rt(nrt)) - goto out_release; + /* No need to remove rt from the exception table if rt is + * a cached route because rt6_insert_exception() will + * takes care of it + */ + if (rt6_insert_exception(nrt, rt)) { + dst_release_immediate(&nrt->dst); + goto out; + } netevent.old = &rt->dst; netevent.new = &nrt->dst; @@ -2942,17 +2970,6 @@ static void rt6_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_bu netevent.neigh = neigh; call_netevent_notifiers(NETEVENT_REDIRECT, &netevent); - if (rt->rt6i_flags & RTF_CACHE) { - rt = (struct rt6_info *) dst_clone(&rt->dst); - ip6_del_rt(rt); - } - -out_release: - /* Release the reference taken in - * ip6_rt_cache_alloc() - */ - dst_release(&nrt->dst); - out: neigh_release(neigh); } @@ -3344,12 +3361,8 @@ static int fib6_clean_tohost(struct rt6_info *rt, void *arg) { struct in6_addr *gateway = (struct in6_addr *)arg; - /* RTF_CACHE_GATEWAY case will be removed once the exception - * table is hooked up to store all cached routes. - */ - if ((((rt->rt6i_flags & RTF_RA_ROUTER) == RTF_RA_ROUTER) || - ((rt->rt6i_flags & RTF_CACHE_GATEWAY) == RTF_CACHE_GATEWAY)) && - ipv6_addr_equal(gateway, &rt->rt6i_gateway)) { + if (((rt->rt6i_flags & RTF_RA_ROUTER) == RTF_RA_ROUTER) && + ipv6_addr_equal(gateway, &rt->rt6i_gateway)) { return -1; } @@ -3438,20 +3451,9 @@ static int rt6_mtu_change_route(struct rt6_info *rt, void *p_arg) dst_metric_raw(&rt->dst, RTAX_MTU) && !dst_metric_locked(&rt->dst, RTAX_MTU)) { spin_lock_bh(&rt6_exception_lock); - /* This case will be removed once the exception table - * is hooked up. - */ - if (rt->rt6i_flags & RTF_CACHE) { - /* For RTF_CACHE with rt6i_pmtu == 0 - * (i.e. a redirected route), - * the metrics of its rt->dst.from has already - * been updated. - */ - if (rt->rt6i_pmtu && rt->rt6i_pmtu > arg->mtu) - rt->rt6i_pmtu = arg->mtu; - } else if (dst_mtu(&rt->dst) >= arg->mtu || - (dst_mtu(&rt->dst) < arg->mtu && - dst_mtu(&rt->dst) == idev->cnf.mtu6)) { + if (dst_mtu(&rt->dst) >= arg->mtu || + (dst_mtu(&rt->dst) < arg->mtu && + dst_mtu(&rt->dst) == idev->cnf.mtu6)) { dst_metric_set(&rt->dst, RTAX_MTU, arg->mtu); } rt6_exceptions_update_pmtu(rt, arg->mtu); -- cgit v1.2.3 From bbd63f06d114a52be33f6982fc89ca2768cdeb62 Mon Sep 17 00:00:00 2001 From: Wei Wang Date: Fri, 6 Oct 2017 12:06:07 -0700 Subject: ipv6: update fn_sernum after route is inserted to tree fib6_add() logic currently calls fib6_add_1() to figure out what node should be used for the newly added route and then call fib6_add_rt2node() to insert the route to the node. And during the call of fib6_add_1(), fn_sernum is updated for all nodes that share the same prefix as the new route. This does not have issue in the current code because reader thread will not be able to access the tree while writer thread is inserting new route to it. However, it is not the case once we transition to use RCU. Reader thread could potentially see the new fn_sernum before the new route is inserted. As a result, reader thread's route lookup will return a stale route with the new fn_sernum. In order to solve this issue, we remove all the update of fn_sernum in fib6_add_1(), and instead, introduce a new function that updates fn_sernum for all related nodes and call this functions once the route is successfully inserted to the tree. Also, smp_wmb() is used after a route is successfully inserted into the fib tree and right before the updated of fn->sernum. And smp_rmb() is used right after fn->sernum is accessed in rt6_get_cookie_safe(). This is to guarantee that when the reader thread sees the new fn->sernum, the new route is already inserted in the tree in memory. Signed-off-by: Wei Wang Signed-off-by: Martin KaFai Lau Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 2 ++ net/ipv6/ip6_fib.c | 39 +++++++++++++++++++++------------------ 2 files changed, 23 insertions(+), 18 deletions(-) (limited to 'include/net') diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index d0b7283073e3..6bf929b50951 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -220,6 +220,8 @@ static inline bool rt6_get_cookie_safe(const struct rt6_info *rt, if (fn) { *cookie = fn->fn_sernum; + /* pairs with smp_wmb() in fib6_update_sernum_upto_root() */ + smp_rmb(); status = true; } diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index eee392f7b1f6..f604b311cc3e 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -585,7 +585,7 @@ out: static struct fib6_node *fib6_add_1(struct fib6_node *root, struct in6_addr *addr, int plen, int offset, int allow_create, - int replace_required, int sernum, + int replace_required, struct netlink_ext_ack *extack) { struct fib6_node *fn, *in, *ln; @@ -631,8 +631,6 @@ static struct fib6_node *fib6_add_1(struct fib6_node *root, fn->leaf = NULL; } - fn->fn_sernum = sernum; - return fn; } @@ -641,7 +639,6 @@ static struct fib6_node *fib6_add_1(struct fib6_node *root, */ /* Try to walk down on tree. */ - fn->fn_sernum = sernum; dir = addr_bit_set(addr, fn->fn_bit); pn = fn; fn = dir ? fn->right : fn->left; @@ -677,7 +674,6 @@ static struct fib6_node *fib6_add_1(struct fib6_node *root, ln->fn_bit = plen; ln->parent = pn; - ln->fn_sernum = sernum; if (dir) pn->right = ln; @@ -737,8 +733,6 @@ insert_above: in->leaf = fn->leaf; atomic_inc(&in->leaf->rt6i_ref); - in->fn_sernum = sernum; - /* update parent pointer */ if (dir) pn->right = in; @@ -750,8 +744,6 @@ insert_above: ln->parent = in; fn->parent = in; - ln->fn_sernum = sernum; - if (addr_bit_set(addr, bit)) { in->right = ln; in->left = fn; @@ -776,8 +768,6 @@ insert_above: ln->parent = pn; - ln->fn_sernum = sernum; - if (dir) pn->right = ln; else @@ -1079,6 +1069,20 @@ void fib6_force_start_gc(struct net *net) jiffies + net->ipv6.sysctl.ip6_rt_gc_interval); } +static void fib6_update_sernum_upto_root(struct rt6_info *rt, + int sernum) +{ + struct fib6_node *fn = rcu_dereference_protected(rt->rt6i_node, + lockdep_is_held(&rt->rt6i_table->tb6_lock)); + + /* paired with smp_rmb() in rt6_get_cookie_safe() */ + smp_wmb(); + while (fn) { + fn->fn_sernum = sernum; + fn = fn->parent; + } +} + /* * Add routing information to the routing tree. * / @@ -1111,7 +1115,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, fn = fib6_add_1(root, &rt->rt6i_dst.addr, rt->rt6i_dst.plen, offsetof(struct rt6_info, rt6i_dst), allow_create, - replace_required, sernum, extack); + replace_required, extack); if (IS_ERR(fn)) { err = PTR_ERR(fn); fn = NULL; @@ -1145,15 +1149,13 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, sfn->leaf = info->nl_net->ipv6.ip6_null_entry; atomic_inc(&info->nl_net->ipv6.ip6_null_entry->rt6i_ref); sfn->fn_flags = RTN_ROOT; - sfn->fn_sernum = sernum; /* Now add the first leaf node to new subtree */ sn = fib6_add_1(sfn, &rt->rt6i_src.addr, rt->rt6i_src.plen, offsetof(struct rt6_info, rt6i_src), - allow_create, replace_required, sernum, - extack); + allow_create, replace_required, extack); if (IS_ERR(sn)) { /* If it is failed, discard just allocated @@ -1172,8 +1174,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, sn = fib6_add_1(fn->subtree, &rt->rt6i_src.addr, rt->rt6i_src.plen, offsetof(struct rt6_info, rt6i_src), - allow_create, replace_required, sernum, - extack); + allow_create, replace_required, extack); if (IS_ERR(sn)) { err = PTR_ERR(sn); @@ -1190,8 +1191,10 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, #endif err = fib6_add_rt2node(fn, rt, info, mxc); - if (!err) + if (!err) { + fib6_update_sernum_upto_root(rt, sernum); fib6_start_gc(info->nl_net, rt); + } out: if (err) { -- cgit v1.2.3 From 66f5d6ce53e665477d2a33e8f539d4fa4ca81c83 Mon Sep 17 00:00:00 2001 From: Wei Wang Date: Fri, 6 Oct 2017 12:06:10 -0700 Subject: ipv6: replace rwlock with rcu and spinlock in fib6_table With all the preparation work before, we are now ready to replace rwlock with rcu and spinlock in fib6_table. That means now all fib6_node in fib6_table are protected by rcu. And when freeing fib6_node, call_rcu() is used to wait for the rcu grace period before releasing the memory. When accessing fib6_node, corresponding rcu APIs need to be used. And all previous sessions protected by the write lock will now be protected by the spin lock per table. All previous sessions protected by read lock will now be protected by rcu_read_lock(). A couple of things to note here: 1. As part of the work of replacing rwlock with rcu, the linked list of fn->leaf now has to be rcu protected as well. So both fn->leaf and rt->dst.rt6_next are now __rcu tagged and corresponding rcu APIs are used when manipulating them. 2. For fn->rr_ptr, first of all, it also needs to be rcu protected now and is tagged with __rcu and rcu APIs are used in corresponding places. Secondly, fn->rr_ptr is changed in rt6_select() which is a reader thread. This makes the issue a bit complicated. We think a valid solution for it is to let rt6_select() grab the tb6_lock if it decides to change it. As it is not in the normal operation and only happens when there is no valid neighbor cache for the route, we think the performance impact should be low. 3. fib6_walk_continue() has to be called with tb6_lock held even in the route dumping related functions, e.g. inet6_dump_fib(), fib6_tables_dump() and ipv6_route_seq_ops. It is because fib6_walk_continue() makes modifications to the walker structure, and so are fib6_repair_tree() and fib6_del_route(). In order to do proper syncing between them, we need to let fib6_walk_continue() hold the lock. We may be able to do further improvement on the way we do the tree walk to get rid of the need for holding the spin lock. But not for now. 4. When fib6_del_route() removes a route from the tree, we no longer mark rt->dst.rt6_next to NULL to make simultaneous reader be able to further traverse the list with rcu. However, rt->dst.rt6_next is only valid within this same rcu period. No one should access it later. 5. All the operation of atomic_inc(rt->rt6i_ref) is changed to be performed before we publish this route (either by linking it to fn->leaf or insert it in the list pointed by fn->leaf) just to be safe because as soon as we publish the route, some read thread will be able to access it. Signed-off-by: Wei Wang Signed-off-by: Martin KaFai Lau Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/dst.h | 2 +- include/net/ip6_fib.h | 24 ++- net/ipv6/addrconf.c | 11 +- net/ipv6/ip6_fib.c | 405 ++++++++++++++++++++++++++++++-------------------- net/ipv6/route.c | 121 ++++++++------- 5 files changed, 333 insertions(+), 230 deletions(-) (limited to 'include/net') diff --git a/include/net/dst.h b/include/net/dst.h index 06a6765da074..204c19e25456 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -101,7 +101,7 @@ struct dst_entry { union { struct dst_entry *next; struct rtable __rcu *rt_next; - struct rt6_info *rt6_next; + struct rt6_info __rcu *rt6_next; struct dn_route __rcu *dn_next; }; }; diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 6bf929b50951..0b438b9bcb10 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -68,18 +68,18 @@ struct fib6_config { }; struct fib6_node { - struct fib6_node *parent; - struct fib6_node *left; - struct fib6_node *right; + struct fib6_node __rcu *parent; + struct fib6_node __rcu *left; + struct fib6_node __rcu *right; #ifdef CONFIG_IPV6_SUBTREES - struct fib6_node *subtree; + struct fib6_node __rcu *subtree; #endif - struct rt6_info *leaf; + struct rt6_info __rcu *leaf; __u16 fn_bit; /* bit key */ __u16 fn_flags; int fn_sernum; - struct rt6_info *rr_ptr; + struct rt6_info __rcu *rr_ptr; struct rcu_head rcu; }; @@ -91,7 +91,7 @@ struct fib6_gc_args { #ifndef CONFIG_IPV6_SUBTREES #define FIB6_SUBTREE(fn) NULL #else -#define FIB6_SUBTREE(fn) ((fn)->subtree) +#define FIB6_SUBTREE(fn) (rcu_dereference_protected((fn)->subtree, 1)) #endif struct mx6_config { @@ -174,6 +174,14 @@ struct rt6_info { unused:7; }; +#define for_each_fib6_node_rt_rcu(fn) \ + for (rt = rcu_dereference((fn)->leaf); rt; \ + rt = rcu_dereference(rt->dst.rt6_next)) + +#define for_each_fib6_walker_rt(w) \ + for (rt = (w)->leaf; rt; \ + rt = rcu_dereference_protected(rt->dst.rt6_next, 1)) + static inline struct inet6_dev *ip6_dst_idev(struct dst_entry *dst) { return ((struct rt6_info *)dst)->rt6i_idev; @@ -310,7 +318,7 @@ struct rt6_statistics { struct fib6_table { struct hlist_node tb6_hlist; u32 tb6_id; - rwlock_t tb6_lock; + spinlock_t tb6_lock; struct fib6_node tb6_root; struct inet_peer_base tb6_peers; unsigned int flags; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index f86e931d555e..9854d93e45bb 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -2321,12 +2321,12 @@ static struct rt6_info *addrconf_get_prefix_route(const struct in6_addr *pfx, if (!table) return NULL; - read_lock_bh(&table->tb6_lock); + rcu_read_lock(); fn = fib6_locate(&table->tb6_root, pfx, plen, NULL, 0, true); if (!fn) goto out; - for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) { + for_each_fib6_node_rt_rcu(fn) { if (rt->dst.dev->ifindex != dev->ifindex) continue; if ((rt->rt6i_flags & flags) != flags) @@ -2338,7 +2338,7 @@ static struct rt6_info *addrconf_get_prefix_route(const struct in6_addr *pfx, break; } out: - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); return rt; } @@ -5898,10 +5898,9 @@ void addrconf_disable_policy_idev(struct inet6_dev *idev, int val) spin_lock(&ifa->lock); if (ifa->rt) { struct rt6_info *rt = ifa->rt; - struct fib6_table *table = rt->rt6i_table; int cpu; - read_lock(&table->tb6_lock); + rcu_read_lock(); addrconf_set_nopolicy(ifa->rt, val); if (rt->rt6i_pcpu) { for_each_possible_cpu(cpu) { @@ -5911,7 +5910,7 @@ void addrconf_disable_policy_idev(struct inet6_dev *idev, int val) addrconf_set_nopolicy(*rtp, val); } } - read_unlock(&table->tb6_lock); + rcu_read_unlock(); } spin_unlock(&ifa->lock); } diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index cf6137e81408..3f95908b39c3 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -54,8 +54,12 @@ struct fib6_cleaner { #define FWS_INIT FWS_L #endif -static struct rt6_info *fib6_find_prefix(struct net *net, struct fib6_node *fn); -static struct fib6_node *fib6_repair_tree(struct net *net, struct fib6_node *fn); +static struct rt6_info *fib6_find_prefix(struct net *net, + struct fib6_table *table, + struct fib6_node *fn); +static struct fib6_node *fib6_repair_tree(struct net *net, + struct fib6_table *table, + struct fib6_node *fn); static int fib6_walk(struct net *net, struct fib6_walker *w); static int fib6_walk_continue(struct fib6_walker *w); @@ -107,12 +111,12 @@ void fib6_update_sernum(struct rt6_info *rt) struct net *net = dev_net(rt->dst.dev); struct fib6_node *fn; - write_lock_bh(&table->tb6_lock); + spin_lock_bh(&table->tb6_lock); fn = rcu_dereference_protected(rt->rt6i_node, lockdep_is_held(&table->tb6_lock)); if (fn) fn->fn_sernum = fib6_new_sernum(net); - write_unlock_bh(&table->tb6_lock); + spin_unlock_bh(&table->tb6_lock); } /* @@ -207,8 +211,7 @@ static void fib6_link_table(struct net *net, struct fib6_table *tb) * Initialize table lock at a single place to give lockdep a key, * tables aren't visible prior to being linked to the list. */ - rwlock_init(&tb->tb6_lock); - + spin_lock_init(&tb->tb6_lock); h = tb->tb6_id & (FIB6_TABLE_HASHSZ - 1); /* @@ -227,7 +230,8 @@ static struct fib6_table *fib6_alloc_table(struct net *net, u32 id) table = kzalloc(sizeof(*table), GFP_ATOMIC); if (table) { table->tb6_id = id; - table->tb6_root.leaf = net->ipv6.ip6_null_entry; + rcu_assign_pointer(table->tb6_root.leaf, + net->ipv6.ip6_null_entry); table->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; inet_peer_base_init(&table->tb6_peers); } @@ -324,11 +328,8 @@ unsigned int fib6_tables_seq_read(struct net *net) struct hlist_head *head = &net->ipv6.fib_table_hash[h]; struct fib6_table *tb; - hlist_for_each_entry_rcu(tb, head, tb6_hlist) { - read_lock_bh(&tb->tb6_lock); + hlist_for_each_entry_rcu(tb, head, tb6_hlist) fib_seq += tb->fib_seq; - read_unlock_bh(&tb->tb6_lock); - } } rcu_read_unlock(); @@ -374,7 +375,7 @@ static int fib6_node_dump(struct fib6_walker *w) { struct rt6_info *rt; - for (rt = w->leaf; rt; rt = rt->dst.rt6_next) + for_each_fib6_walker_rt(w) fib6_rt_dump(rt, w->args); w->leaf = NULL; return 0; @@ -384,9 +385,9 @@ static void fib6_table_dump(struct net *net, struct fib6_table *tb, struct fib6_walker *w) { w->root = &tb->tb6_root; - read_lock_bh(&tb->tb6_lock); + spin_lock_bh(&tb->tb6_lock); fib6_walk(net, w); - read_unlock_bh(&tb->tb6_lock); + spin_unlock_bh(&tb->tb6_lock); } /* Called with rcu_read_lock() */ @@ -423,7 +424,7 @@ static int fib6_dump_node(struct fib6_walker *w) int res; struct rt6_info *rt; - for (rt = w->leaf; rt; rt = rt->dst.rt6_next) { + for_each_fib6_walker_rt(w) { res = rt6_dump_route(rt, w->args); if (res < 0) { /* Frame is full, suspend walking */ @@ -482,9 +483,9 @@ static int fib6_dump_table(struct fib6_table *table, struct sk_buff *skb, w->count = 0; w->skip = 0; - read_lock_bh(&table->tb6_lock); + spin_lock_bh(&table->tb6_lock); res = fib6_walk(net, w); - read_unlock_bh(&table->tb6_lock); + spin_unlock_bh(&table->tb6_lock); if (res > 0) { cb->args[4] = 1; cb->args[5] = w->root->fn_sernum; @@ -499,9 +500,9 @@ static int fib6_dump_table(struct fib6_table *table, struct sk_buff *skb, } else w->skip = 0; - read_lock_bh(&table->tb6_lock); + spin_lock_bh(&table->tb6_lock); res = fib6_walk_continue(w); - read_unlock_bh(&table->tb6_lock); + spin_unlock_bh(&table->tb6_lock); if (res <= 0) { fib6_walker_unlink(net, w); cb->args[4] = 0; @@ -582,11 +583,12 @@ out: * node. */ -static struct fib6_node *fib6_add_1(struct fib6_node *root, - struct in6_addr *addr, int plen, - int offset, int allow_create, - int replace_required, - struct netlink_ext_ack *extack) +static struct fib6_node *fib6_add_1(struct fib6_table *table, + struct fib6_node *root, + struct in6_addr *addr, int plen, + int offset, int allow_create, + int replace_required, + struct netlink_ext_ack *extack) { struct fib6_node *fn, *in, *ln; struct fib6_node *pn = NULL; @@ -601,7 +603,9 @@ static struct fib6_node *fib6_add_1(struct fib6_node *root, fn = root; do { - key = (struct rt6key *)((u8 *)fn->leaf + offset); + struct rt6_info *leaf = rcu_dereference_protected(fn->leaf, + lockdep_is_held(&table->tb6_lock)); + key = (struct rt6key *)((u8 *)leaf + offset); /* * Prefix match @@ -627,8 +631,8 @@ static struct fib6_node *fib6_add_1(struct fib6_node *root, if (plen == fn->fn_bit) { /* clean up an intermediate node */ if (!(fn->fn_flags & RTN_RTINFO)) { - rt6_release(fn->leaf); - fn->leaf = NULL; + RCU_INIT_POINTER(fn->leaf, NULL); + rt6_release(leaf); } return fn; @@ -641,7 +645,11 @@ static struct fib6_node *fib6_add_1(struct fib6_node *root, /* Try to walk down on tree. */ dir = addr_bit_set(addr, fn->fn_bit); pn = fn; - fn = dir ? fn->right : fn->left; + fn = dir ? + rcu_dereference_protected(fn->right, + lockdep_is_held(&table->tb6_lock)) : + rcu_dereference_protected(fn->left, + lockdep_is_held(&table->tb6_lock)); } while (fn); if (!allow_create) { @@ -672,13 +680,12 @@ static struct fib6_node *fib6_add_1(struct fib6_node *root, if (!ln) return ERR_PTR(-ENOMEM); ln->fn_bit = plen; - - ln->parent = pn; + RCU_INIT_POINTER(ln->parent, pn); if (dir) - pn->right = ln; + rcu_assign_pointer(pn->right, ln); else - pn->left = ln; + rcu_assign_pointer(pn->left, ln); return ln; @@ -692,7 +699,8 @@ insert_above: * and the current */ - pn = fn->parent; + pn = rcu_dereference_protected(fn->parent, + lockdep_is_held(&table->tb6_lock)); /* find 1st bit in difference between the 2 addrs. @@ -729,27 +737,28 @@ insert_above: in->fn_bit = bit; - in->parent = pn; + RCU_INIT_POINTER(in->parent, pn); in->leaf = fn->leaf; - atomic_inc(&in->leaf->rt6i_ref); + atomic_inc(&rcu_dereference_protected(in->leaf, + lockdep_is_held(&table->tb6_lock))->rt6i_ref); /* update parent pointer */ if (dir) - pn->right = in; + rcu_assign_pointer(pn->right, in); else - pn->left = in; + rcu_assign_pointer(pn->left, in); ln->fn_bit = plen; - ln->parent = in; - fn->parent = in; + RCU_INIT_POINTER(ln->parent, in); + rcu_assign_pointer(fn->parent, in); if (addr_bit_set(addr, bit)) { - in->right = ln; - in->left = fn; + rcu_assign_pointer(in->right, ln); + rcu_assign_pointer(in->left, fn); } else { - in->left = ln; - in->right = fn; + rcu_assign_pointer(in->left, ln); + rcu_assign_pointer(in->right, fn); } } else { /* plen <= bit */ @@ -766,19 +775,19 @@ insert_above: ln->fn_bit = plen; - ln->parent = pn; - - if (dir) - pn->right = ln; - else - pn->left = ln; + RCU_INIT_POINTER(ln->parent, pn); if (addr_bit_set(&key->addr, plen)) - ln->right = fn; + RCU_INIT_POINTER(ln->right, fn); else - ln->left = fn; + RCU_INIT_POINTER(ln->left, fn); + + rcu_assign_pointer(fn->parent, ln); - fn->parent = ln; + if (dir) + rcu_assign_pointer(pn->right, ln); + else + rcu_assign_pointer(pn->left, ln); } return ln; } @@ -824,6 +833,8 @@ static int fib6_commit_metrics(struct dst_entry *dst, struct mx6_config *mxc) static void fib6_purge_rt(struct rt6_info *rt, struct fib6_node *fn, struct net *net) { + struct fib6_table *table = rt->rt6i_table; + if (atomic_read(&rt->rt6i_ref) != 1) { /* This route is used as dummy address holder in some split * nodes. It is not leaked, but it still holds other resources, @@ -832,12 +843,17 @@ static void fib6_purge_rt(struct rt6_info *rt, struct fib6_node *fn, * to still alive ones. */ while (fn) { - if (!(fn->fn_flags & RTN_RTINFO) && fn->leaf == rt) { - fn->leaf = fib6_find_prefix(net, fn); - atomic_inc(&fn->leaf->rt6i_ref); + struct rt6_info *leaf = rcu_dereference_protected(fn->leaf, + lockdep_is_held(&table->tb6_lock)); + struct rt6_info *new_leaf; + if (!(fn->fn_flags & RTN_RTINFO) && leaf == rt) { + new_leaf = fib6_find_prefix(net, table, fn); + atomic_inc(&new_leaf->rt6i_ref); + rcu_assign_pointer(fn->leaf, new_leaf); rt6_release(rt); } - fn = fn->parent; + fn = rcu_dereference_protected(fn->parent, + lockdep_is_held(&table->tb6_lock)); } } } @@ -849,9 +865,11 @@ static void fib6_purge_rt(struct rt6_info *rt, struct fib6_node *fn, static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt, struct nl_info *info, struct mx6_config *mxc) { + struct rt6_info *leaf = rcu_dereference_protected(fn->leaf, + lockdep_is_held(&rt->rt6i_table->tb6_lock)); struct rt6_info *iter = NULL; - struct rt6_info **ins; - struct rt6_info **fallback_ins = NULL; + struct rt6_info __rcu **ins; + struct rt6_info __rcu **fallback_ins = NULL; int replace = (info->nlh && (info->nlh->nlmsg_flags & NLM_F_REPLACE)); int add = (!info->nlh || @@ -866,7 +884,9 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt, ins = &fn->leaf; - for (iter = fn->leaf; iter; iter = iter->dst.rt6_next) { + for (iter = leaf; iter; + iter = rcu_dereference_protected(iter->dst.rt6_next, + lockdep_is_held(&rt->rt6i_table->tb6_lock))) { /* * Search for duplicates */ @@ -928,7 +948,8 @@ next_iter: if (fallback_ins && !found) { /* No ECMP-able route found, replace first non-ECMP one */ ins = fallback_ins; - iter = *ins; + iter = rcu_dereference_protected(*ins, + lockdep_is_held(&rt->rt6i_table->tb6_lock)); found++; } @@ -942,7 +963,7 @@ next_iter: struct rt6_info *sibling, *temp_sibling; /* Find the first route that have the same metric */ - sibling = fn->leaf; + sibling = leaf; while (sibling) { if (sibling->rt6i_metric == rt->rt6i_metric && rt6_qualify_for_ecmp(sibling)) { @@ -950,7 +971,8 @@ next_iter: &sibling->rt6i_siblings); break; } - sibling = sibling->dst.rt6_next; + sibling = rcu_dereference_protected(sibling->dst.rt6_next, + lockdep_is_held(&rt->rt6i_table->tb6_lock)); } /* For each sibling in the list, increment the counter of * siblings. BUG() if counters does not match, list of siblings @@ -979,10 +1001,10 @@ add: if (err) return err; - rt->dst.rt6_next = iter; - *ins = rt; - rcu_assign_pointer(rt->rt6i_node, fn); + rcu_assign_pointer(rt->dst.rt6_next, iter); atomic_inc(&rt->rt6i_ref); + rcu_assign_pointer(rt->rt6i_node, fn); + rcu_assign_pointer(*ins, rt); call_fib6_entry_notifiers(info->nl_net, FIB_EVENT_ENTRY_ADD, rt); if (!info->skip_notify) @@ -1008,10 +1030,10 @@ add: if (err) return err; - *ins = rt; + atomic_inc(&rt->rt6i_ref); rcu_assign_pointer(rt->rt6i_node, fn); rt->dst.rt6_next = iter->dst.rt6_next; - atomic_inc(&rt->rt6i_ref); + rcu_assign_pointer(*ins, rt); call_fib6_entry_notifiers(info->nl_net, FIB_EVENT_ENTRY_REPLACE, rt); if (!info->skip_notify) @@ -1023,14 +1045,15 @@ add: nsiblings = iter->rt6i_nsiblings; iter->rt6i_node = NULL; fib6_purge_rt(iter, fn, info->nl_net); - if (fn->rr_ptr == iter) + if (rcu_access_pointer(fn->rr_ptr) == iter) fn->rr_ptr = NULL; rt6_release(iter); if (nsiblings) { /* Replacing an ECMP route, remove all siblings */ ins = &rt->dst.rt6_next; - iter = *ins; + iter = rcu_dereference_protected(*ins, + lockdep_is_held(&rt->rt6i_table->tb6_lock)); while (iter) { if (iter->rt6i_metric > rt->rt6i_metric) break; @@ -1038,14 +1061,15 @@ add: *ins = iter->dst.rt6_next; iter->rt6i_node = NULL; fib6_purge_rt(iter, fn, info->nl_net); - if (fn->rr_ptr == iter) + if (rcu_access_pointer(fn->rr_ptr) == iter) fn->rr_ptr = NULL; rt6_release(iter); nsiblings--; } else { ins = &iter->dst.rt6_next; } - iter = *ins; + iter = rcu_dereference_protected(*ins, + lockdep_is_held(&rt->rt6i_table->tb6_lock)); } WARN_ON(nsiblings != 0); } @@ -1079,7 +1103,8 @@ static void fib6_update_sernum_upto_root(struct rt6_info *rt, smp_wmb(); while (fn) { fn->fn_sernum = sernum; - fn = fn->parent; + fn = rcu_dereference_protected(fn->parent, + lockdep_is_held(&rt->rt6i_table->tb6_lock)); } } @@ -1087,12 +1112,14 @@ static void fib6_update_sernum_upto_root(struct rt6_info *rt, * Add routing information to the routing tree. * / * with source addr info in sub-trees + * Need to own table->tb6_lock */ int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nl_info *info, struct mx6_config *mxc, struct netlink_ext_ack *extack) { + struct fib6_table *table = rt->rt6i_table; struct fib6_node *fn, *pn = NULL; int err = -ENOMEM; int allow_create = 1; @@ -1113,7 +1140,8 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, if (!allow_create && !replace_required) pr_warn("RTM_NEWROUTE with no NLM_F_CREATE or NLM_F_REPLACE\n"); - fn = fib6_add_1(root, &rt->rt6i_dst.addr, rt->rt6i_dst.plen, + fn = fib6_add_1(table, root, + &rt->rt6i_dst.addr, rt->rt6i_dst.plen, offsetof(struct rt6_info, rt6i_dst), allow_create, replace_required, extack); if (IS_ERR(fn)) { @@ -1128,7 +1156,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, if (rt->rt6i_src.plen) { struct fib6_node *sn; - if (!fn->subtree) { + if (!rcu_access_pointer(fn->subtree)) { struct fib6_node *sfn; /* @@ -1146,13 +1174,14 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, if (!sfn) goto failure; - sfn->leaf = info->nl_net->ipv6.ip6_null_entry; atomic_inc(&info->nl_net->ipv6.ip6_null_entry->rt6i_ref); + rcu_assign_pointer(sfn->leaf, + info->nl_net->ipv6.ip6_null_entry); sfn->fn_flags = RTN_ROOT; /* Now add the first leaf node to new subtree */ - sn = fib6_add_1(sfn, &rt->rt6i_src.addr, + sn = fib6_add_1(table, sfn, &rt->rt6i_src.addr, rt->rt6i_src.plen, offsetof(struct rt6_info, rt6i_src), allow_create, replace_required, extack); @@ -1168,10 +1197,10 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, } /* Now link new subtree to main tree */ - sfn->parent = fn; - fn->subtree = sfn; + rcu_assign_pointer(sfn->parent, fn); + rcu_assign_pointer(fn->subtree, sfn); } else { - sn = fib6_add_1(fn->subtree, &rt->rt6i_src.addr, + sn = fib6_add_1(table, FIB6_SUBTREE(fn), &rt->rt6i_src.addr, rt->rt6i_src.plen, offsetof(struct rt6_info, rt6i_src), allow_create, replace_required, extack); @@ -1182,9 +1211,9 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, } } - if (!fn->leaf) { - fn->leaf = rt; + if (!rcu_access_pointer(fn->leaf)) { atomic_inc(&rt->rt6i_ref); + rcu_assign_pointer(fn->leaf, rt); } fn = sn; } @@ -1203,19 +1232,23 @@ out: * If fib6_add_1 has cleared the old leaf pointer in the * super-tree leaf node we have to find a new one for it. */ - if (pn != fn && pn->leaf == rt) { - pn->leaf = NULL; + struct rt6_info *pn_leaf = rcu_dereference_protected(pn->leaf, + lockdep_is_held(&table->tb6_lock)); + if (pn != fn && pn_leaf == rt) { + pn_leaf = NULL; + RCU_INIT_POINTER(pn->leaf, NULL); atomic_dec(&rt->rt6i_ref); } - if (pn != fn && !pn->leaf && !(pn->fn_flags & RTN_RTINFO)) { - pn->leaf = fib6_find_prefix(info->nl_net, pn); + if (pn != fn && !pn_leaf && !(pn->fn_flags & RTN_RTINFO)) { + pn_leaf = fib6_find_prefix(info->nl_net, table, pn); #if RT6_DEBUG >= 2 - if (!pn->leaf) { - WARN_ON(pn->leaf == NULL); - pn->leaf = info->nl_net->ipv6.ip6_null_entry; + if (!pn_leaf) { + WARN_ON(!pn_leaf); + pn_leaf = info->nl_net->ipv6.ip6_null_entry; } #endif - atomic_inc(&pn->leaf->rt6i_ref); + atomic_inc(&pn_leaf->rt6i_ref); + rcu_assign_pointer(pn->leaf, pn_leaf); } #endif goto failure; @@ -1230,7 +1263,7 @@ failure: * fn->leaf. */ if (fn && !(fn->fn_flags & (RTN_RTINFO|RTN_ROOT))) - fib6_repair_tree(info->nl_net, fn); + fib6_repair_tree(info->nl_net, table, fn); /* Always release dst as dst->__refcnt is guaranteed * to be taken before entering this function */ @@ -1268,7 +1301,8 @@ static struct fib6_node *fib6_lookup_1(struct fib6_node *root, dir = addr_bit_set(args->addr, fn->fn_bit); - next = dir ? fn->right : fn->left; + next = dir ? rcu_dereference(fn->right) : + rcu_dereference(fn->left); if (next) { fn = next; @@ -1278,8 +1312,10 @@ static struct fib6_node *fib6_lookup_1(struct fib6_node *root, } while (fn) { - if (FIB6_SUBTREE(fn) || fn->fn_flags & RTN_RTINFO) { - struct rt6_info *leaf = fn->leaf; + struct fib6_node *subtree = FIB6_SUBTREE(fn); + + if (subtree || fn->fn_flags & RTN_RTINFO) { + struct rt6_info *leaf = rcu_dereference(fn->leaf); struct rt6key *key; if (!leaf) @@ -1289,10 +1325,9 @@ static struct fib6_node *fib6_lookup_1(struct fib6_node *root, if (ipv6_prefix_equal(&key->addr, args->addr, key->plen)) { #ifdef CONFIG_IPV6_SUBTREES - if (fn->subtree) { + if (subtree) { struct fib6_node *sfn; - sfn = fib6_lookup_1(fn->subtree, - args + 1); + sfn = fib6_lookup_1(subtree, args + 1); if (!sfn) goto backtrack; fn = sfn; @@ -1306,12 +1341,14 @@ backtrack: if (fn->fn_flags & RTN_ROOT) break; - fn = fn->parent; + fn = rcu_dereference(fn->parent); } return NULL; } +/* called with rcu_read_lock() held + */ struct fib6_node *fib6_lookup(struct fib6_node *root, const struct in6_addr *daddr, const struct in6_addr *saddr) { @@ -1359,7 +1396,7 @@ static struct fib6_node *fib6_locate_1(struct fib6_node *root, struct fib6_node *fn, *prev = NULL; for (fn = root; fn ; ) { - struct rt6_info *leaf = fn->leaf; + struct rt6_info *leaf = rcu_dereference(fn->leaf); struct rt6key *key; /* This node is being deleted */ @@ -1389,9 +1426,9 @@ next: * We have more bits to go */ if (addr_bit_set(addr, fn->fn_bit)) - fn = fn->right; + fn = rcu_dereference(fn->right); else - fn = fn->left; + fn = rcu_dereference(fn->left); } out: if (exact_match) @@ -1413,9 +1450,11 @@ struct fib6_node *fib6_locate(struct fib6_node *root, #ifdef CONFIG_IPV6_SUBTREES if (src_len) { + struct fib6_node *subtree = FIB6_SUBTREE(fn); + WARN_ON(saddr == NULL); - if (fn && fn->subtree) - fn = fib6_locate_1(fn->subtree, saddr, src_len, + if (fn && subtree) + fn = fib6_locate_1(subtree, saddr, src_len, offsetof(struct rt6_info, rt6i_src), exact_match); } @@ -1433,16 +1472,26 @@ struct fib6_node *fib6_locate(struct fib6_node *root, * */ -static struct rt6_info *fib6_find_prefix(struct net *net, struct fib6_node *fn) +static struct rt6_info *fib6_find_prefix(struct net *net, + struct fib6_table *table, + struct fib6_node *fn) { + struct fib6_node *child_left, *child_right; + if (fn->fn_flags & RTN_ROOT) return net->ipv6.ip6_null_entry; while (fn) { - if (fn->left) - return fn->left->leaf; - if (fn->right) - return fn->right->leaf; + child_left = rcu_dereference_protected(fn->left, + lockdep_is_held(&table->tb6_lock)); + child_right = rcu_dereference_protected(fn->right, + lockdep_is_held(&table->tb6_lock)); + if (child_left) + return rcu_dereference_protected(child_left->leaf, + lockdep_is_held(&table->tb6_lock)); + if (child_right) + return rcu_dereference_protected(child_right->leaf, + lockdep_is_held(&table->tb6_lock)); fn = FIB6_SUBTREE(fn); } @@ -1452,31 +1501,49 @@ static struct rt6_info *fib6_find_prefix(struct net *net, struct fib6_node *fn) /* * Called to trim the tree of intermediate nodes when possible. "fn" * is the node we want to try and remove. + * Need to own table->tb6_lock */ static struct fib6_node *fib6_repair_tree(struct net *net, - struct fib6_node *fn) + struct fib6_table *table, + struct fib6_node *fn) { int children; int nstate; - struct fib6_node *child, *pn; + struct fib6_node *child; struct fib6_walker *w; int iter = 0; for (;;) { + struct fib6_node *fn_r = rcu_dereference_protected(fn->right, + lockdep_is_held(&table->tb6_lock)); + struct fib6_node *fn_l = rcu_dereference_protected(fn->left, + lockdep_is_held(&table->tb6_lock)); + struct fib6_node *pn = rcu_dereference_protected(fn->parent, + lockdep_is_held(&table->tb6_lock)); + struct fib6_node *pn_r = rcu_dereference_protected(pn->right, + lockdep_is_held(&table->tb6_lock)); + struct fib6_node *pn_l = rcu_dereference_protected(pn->left, + lockdep_is_held(&table->tb6_lock)); + struct rt6_info *fn_leaf = rcu_dereference_protected(fn->leaf, + lockdep_is_held(&table->tb6_lock)); + struct rt6_info *pn_leaf = rcu_dereference_protected(pn->leaf, + lockdep_is_held(&table->tb6_lock)); + struct rt6_info *new_fn_leaf; + RT6_TRACE("fixing tree: plen=%d iter=%d\n", fn->fn_bit, iter); iter++; WARN_ON(fn->fn_flags & RTN_RTINFO); WARN_ON(fn->fn_flags & RTN_TL_ROOT); - WARN_ON(fn->leaf); + WARN_ON(fn_leaf); children = 0; child = NULL; - if (fn->right) - child = fn->right, children |= 1; - if (fn->left) - child = fn->left, children |= 2; + if (fn_r) + child = fn_r, children |= 1; + if (fn_l) + child = fn_l, children |= 2; if (children == 3 || FIB6_SUBTREE(fn) #ifdef CONFIG_IPV6_SUBTREES @@ -1484,36 +1551,36 @@ static struct fib6_node *fib6_repair_tree(struct net *net, || (children && fn->fn_flags & RTN_ROOT) #endif ) { - fn->leaf = fib6_find_prefix(net, fn); + new_fn_leaf = fib6_find_prefix(net, table, fn); #if RT6_DEBUG >= 2 - if (!fn->leaf) { - WARN_ON(!fn->leaf); - fn->leaf = net->ipv6.ip6_null_entry; + if (!new_fn_leaf) { + WARN_ON(!new_fn_leaf); + new_fn_leaf = net->ipv6.ip6_null_entry; } #endif - atomic_inc(&fn->leaf->rt6i_ref); - return fn->parent; + atomic_inc(&new_fn_leaf->rt6i_ref); + rcu_assign_pointer(fn->leaf, new_fn_leaf); + return pn; } - pn = fn->parent; #ifdef CONFIG_IPV6_SUBTREES if (FIB6_SUBTREE(pn) == fn) { WARN_ON(!(fn->fn_flags & RTN_ROOT)); - FIB6_SUBTREE(pn) = NULL; + RCU_INIT_POINTER(pn->subtree, NULL); nstate = FWS_L; } else { WARN_ON(fn->fn_flags & RTN_ROOT); #endif - if (pn->right == fn) - pn->right = child; - else if (pn->left == fn) - pn->left = child; + if (pn_r == fn) + rcu_assign_pointer(pn->right, child); + else if (pn_l == fn) + rcu_assign_pointer(pn->left, child); #if RT6_DEBUG >= 2 else WARN_ON(1); #endif if (child) - child->parent = pn; + rcu_assign_pointer(child->parent, pn); nstate = FWS_R; #ifdef CONFIG_IPV6_SUBTREES } @@ -1546,17 +1613,18 @@ static struct fib6_node *fib6_repair_tree(struct net *net, if (pn->fn_flags & RTN_RTINFO || FIB6_SUBTREE(pn)) return pn; - rt6_release(pn->leaf); - pn->leaf = NULL; + RCU_INIT_POINTER(pn->leaf, NULL); + rt6_release(pn_leaf); fn = pn; } } -static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp, - struct nl_info *info) +static void fib6_del_route(struct fib6_table *table, struct fib6_node *fn, + struct rt6_info __rcu **rtp, struct nl_info *info) { struct fib6_walker *w; - struct rt6_info *rt = *rtp; + struct rt6_info *rt = rcu_dereference_protected(*rtp, + lockdep_is_held(&table->tb6_lock)); struct net *net = info->nl_net; RT6_TRACE("fib6_del_route\n"); @@ -1573,7 +1641,7 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp, rt6_flush_exceptions(rt); /* Reset round-robin state, if necessary */ - if (fn->rr_ptr == rt) + if (rcu_access_pointer(fn->rr_ptr) == rt) fn->rr_ptr = NULL; /* Remove this entry from other siblings */ @@ -1592,20 +1660,19 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp, FOR_WALKERS(net, w) { if (w->state == FWS_C && w->leaf == rt) { RT6_TRACE("walker %p adjusted by delroute\n", w); - w->leaf = rt->dst.rt6_next; + w->leaf = rcu_dereference_protected(rt->dst.rt6_next, + lockdep_is_held(&table->tb6_lock)); if (!w->leaf) w->state = FWS_U; } } read_unlock(&net->ipv6.fib6_walker_lock); - rt->dst.rt6_next = NULL; - /* If it was last route, expunge its radix tree node */ - if (!fn->leaf) { + if (!rcu_access_pointer(fn->leaf)) { fn->fn_flags &= ~RTN_RTINFO; net->ipv6.rt6_stats->fib_route_nodes--; - fn = fib6_repair_tree(net, fn); + fn = fib6_repair_tree(net, table, fn); } fib6_purge_rt(rt, fn, net); @@ -1616,12 +1683,15 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp, rt6_release(rt); } +/* Need to own table->tb6_lock */ int fib6_del(struct rt6_info *rt, struct nl_info *info) { struct fib6_node *fn = rcu_dereference_protected(rt->rt6i_node, lockdep_is_held(&rt->rt6i_table->tb6_lock)); + struct fib6_table *table = rt->rt6i_table; struct net *net = info->nl_net; - struct rt6_info **rtp; + struct rt6_info __rcu **rtp; + struct rt6_info __rcu **rtp_next; #if RT6_DEBUG >= 2 if (rt->dst.obsolete > 0) { @@ -1642,11 +1712,14 @@ int fib6_del(struct rt6_info *rt, struct nl_info *info) * Walk the leaf entries looking for ourself */ - for (rtp = &fn->leaf; *rtp; rtp = &(*rtp)->dst.rt6_next) { - if (*rtp == rt) { - fib6_del_route(fn, rtp, info); + for (rtp = &fn->leaf; *rtp; rtp = rtp_next) { + struct rt6_info *cur = rcu_dereference_protected(*rtp, + lockdep_is_held(&table->tb6_lock)); + if (rt == cur) { + fib6_del_route(table, fn, rtp, info); return 0; } + rtp_next = &cur->dst.rt6_next; } return -ENOENT; } @@ -1673,11 +1746,13 @@ int fib6_del(struct rt6_info *rt, struct nl_info *info) * 0 -> walk is complete. * >0 -> walk is incomplete (i.e. suspended) * <0 -> walk is terminated by an error. + * + * This function is called with tb6_lock held. */ static int fib6_walk_continue(struct fib6_walker *w) { - struct fib6_node *fn, *pn; + struct fib6_node *fn, *pn, *left, *right; /* w->root should always be table->tb6_root */ WARN_ON_ONCE(!(w->root->fn_flags & RTN_TL_ROOT)); @@ -1697,20 +1772,22 @@ static int fib6_walk_continue(struct fib6_walker *w) w->state = FWS_L; #endif case FWS_L: - if (fn->left) { - w->node = fn->left; + left = rcu_dereference_protected(fn->left, 1); + if (left) { + w->node = left; w->state = FWS_INIT; continue; } w->state = FWS_R; case FWS_R: - if (fn->right) { - w->node = fn->right; + right = rcu_dereference_protected(fn->right, 1); + if (right) { + w->node = right; w->state = FWS_INIT; continue; } w->state = FWS_C; - w->leaf = fn->leaf; + w->leaf = rcu_dereference_protected(fn->leaf, 1); case FWS_C: if (w->leaf && fn->fn_flags & RTN_RTINFO) { int err; @@ -1732,7 +1809,9 @@ skip: case FWS_U: if (fn == w->root) return 0; - pn = fn->parent; + pn = rcu_dereference_protected(fn->parent, 1); + left = rcu_dereference_protected(pn->left, 1); + right = rcu_dereference_protected(pn->right, 1); w->node = pn; #ifdef CONFIG_IPV6_SUBTREES if (FIB6_SUBTREE(pn) == fn) { @@ -1741,13 +1820,13 @@ skip: continue; } #endif - if (pn->left == fn) { + if (left == fn) { w->state = FWS_R; continue; } - if (pn->right == fn) { + if (right == fn) { w->state = FWS_C; - w->leaf = w->node->leaf; + w->leaf = rcu_dereference_protected(w->node->leaf, 1); continue; } #if RT6_DEBUG >= 2 @@ -1790,7 +1869,7 @@ static int fib6_clean_node(struct fib6_walker *w) return 0; } - for (rt = w->leaf; rt; rt = rt->dst.rt6_next) { + for_each_fib6_walker_rt(w) { res = c->func(rt, c->arg); if (res < 0) { w->leaf = rt; @@ -1850,10 +1929,10 @@ static void __fib6_clean_all(struct net *net, for (h = 0; h < FIB6_TABLE_HASHSZ; h++) { head = &net->ipv6.fib_table_hash[h]; hlist_for_each_entry_rcu(table, head, tb6_hlist) { - write_lock_bh(&table->tb6_lock); + spin_lock_bh(&table->tb6_lock); fib6_clean_tree(net, &table->tb6_root, func, sernum, arg); - write_unlock_bh(&table->tb6_lock); + spin_unlock_bh(&table->tb6_lock); } } rcu_read_unlock(); @@ -1967,7 +2046,8 @@ static int __net_init fib6_net_init(struct net *net) goto out_fib_table_hash; net->ipv6.fib6_main_tbl->tb6_id = RT6_TABLE_MAIN; - net->ipv6.fib6_main_tbl->tb6_root.leaf = net->ipv6.ip6_null_entry; + rcu_assign_pointer(net->ipv6.fib6_main_tbl->tb6_root.leaf, + net->ipv6.ip6_null_entry); net->ipv6.fib6_main_tbl->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; inet_peer_base_init(&net->ipv6.fib6_main_tbl->tb6_peers); @@ -1978,7 +2058,8 @@ static int __net_init fib6_net_init(struct net *net) if (!net->ipv6.fib6_local_tbl) goto out_fib6_main_tbl; net->ipv6.fib6_local_tbl->tb6_id = RT6_TABLE_LOCAL; - net->ipv6.fib6_local_tbl->tb6_root.leaf = net->ipv6.ip6_null_entry; + rcu_assign_pointer(net->ipv6.fib6_local_tbl->tb6_root.leaf, + net->ipv6.ip6_null_entry); net->ipv6.fib6_local_tbl->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; inet_peer_base_init(&net->ipv6.fib6_local_tbl->tb6_peers); @@ -2108,7 +2189,9 @@ static int ipv6_route_yield(struct fib6_walker *w) return 1; do { - iter->w.leaf = iter->w.leaf->dst.rt6_next; + iter->w.leaf = rcu_dereference_protected( + iter->w.leaf->dst.rt6_next, + lockdep_is_held(&iter->tbl->tb6_lock)); iter->skip--; if (!iter->skip && iter->w.leaf) return 1; @@ -2173,7 +2256,7 @@ static void *ipv6_route_seq_next(struct seq_file *seq, void *v, loff_t *pos) if (!v) goto iter_table; - n = ((struct rt6_info *)v)->dst.rt6_next; + n = rcu_dereference(((struct rt6_info *)v)->dst.rt6_next); if (n) { ++*pos; return n; @@ -2181,9 +2264,9 @@ static void *ipv6_route_seq_next(struct seq_file *seq, void *v, loff_t *pos) iter_table: ipv6_route_check_sernum(iter); - read_lock(&iter->tbl->tb6_lock); + spin_lock_bh(&iter->tbl->tb6_lock); r = fib6_walk_continue(&iter->w); - read_unlock(&iter->tbl->tb6_lock); + spin_unlock_bh(&iter->tbl->tb6_lock); if (r > 0) { if (v) ++*pos; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 24b80f43bbfb..cf44d0994b1e 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -488,7 +488,7 @@ static struct rt6_info *rt6_multipath_select(struct rt6_info *match, } /* - * Route lookup. Any table->tb6_lock is implied. + * Route lookup. rcu_read_lock() should be held. */ static inline struct rt6_info *rt6_device_match(struct net *net, @@ -503,7 +503,7 @@ static inline struct rt6_info *rt6_device_match(struct net *net, if (!oif && ipv6_addr_any(saddr)) goto out; - for (sprt = rt; sprt; sprt = sprt->dst.rt6_next) { + for (sprt = rt; sprt; sprt = rcu_dereference(sprt->dst.rt6_next)) { struct net_device *dev = sprt->dst.dev; if (oif) { @@ -722,7 +722,7 @@ static struct rt6_info *find_rr_leaf(struct fib6_node *fn, match = NULL; cont = NULL; - for (rt = rr_head; rt; rt = rt->dst.rt6_next) { + for (rt = rr_head; rt; rt = rcu_dereference(rt->dst.rt6_next)) { if (rt->rt6i_metric != metric) { cont = rt; break; @@ -731,7 +731,8 @@ static struct rt6_info *find_rr_leaf(struct fib6_node *fn, match = find_match(rt, oif, strict, &mpri, match, do_rr); } - for (rt = leaf; rt && rt != rr_head; rt = rt->dst.rt6_next) { + for (rt = leaf; rt && rt != rr_head; + rt = rcu_dereference(rt->dst.rt6_next)) { if (rt->rt6i_metric != metric) { cont = rt; break; @@ -743,7 +744,7 @@ static struct rt6_info *find_rr_leaf(struct fib6_node *fn, if (match || !cont) return match; - for (rt = cont; rt; rt = rt->dst.rt6_next) + for (rt = cont; rt; rt = rcu_dereference(rt->dst.rt6_next)) match = find_match(rt, oif, strict, &mpri, match, do_rr); return match; @@ -752,7 +753,7 @@ static struct rt6_info *find_rr_leaf(struct fib6_node *fn, static struct rt6_info *rt6_select(struct net *net, struct fib6_node *fn, int oif, int strict) { - struct rt6_info *leaf = fn->leaf; + struct rt6_info *leaf = rcu_dereference(fn->leaf); struct rt6_info *match, *rt0; bool do_rr = false; int key_plen; @@ -760,9 +761,9 @@ static struct rt6_info *rt6_select(struct net *net, struct fib6_node *fn, if (!leaf) return net->ipv6.ip6_null_entry; - rt0 = fn->rr_ptr; + rt0 = rcu_dereference(fn->rr_ptr); if (!rt0) - fn->rr_ptr = rt0 = leaf; + rt0 = leaf; /* Double check to make sure fn is not an intermediate node * and fn->leaf does not points to its child's leaf @@ -781,14 +782,19 @@ static struct rt6_info *rt6_select(struct net *net, struct fib6_node *fn, &do_rr); if (do_rr) { - struct rt6_info *next = rt0->dst.rt6_next; + struct rt6_info *next = rcu_dereference(rt0->dst.rt6_next); /* no entries matched; do round-robin */ if (!next || next->rt6i_metric != rt0->rt6i_metric) next = leaf; - if (next != rt0) - fn->rr_ptr = next; + if (next != rt0) { + spin_lock_bh(&leaf->rt6i_table->tb6_lock); + /* make sure next is not being deleted from the tree */ + if (next->rt6i_node) + rcu_assign_pointer(fn->rr_ptr, next); + spin_unlock_bh(&leaf->rt6i_table->tb6_lock); + } } return match ? match : net->ipv6.ip6_null_entry; @@ -878,13 +884,14 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len, static struct fib6_node* fib6_backtrack(struct fib6_node *fn, struct in6_addr *saddr) { - struct fib6_node *pn; + struct fib6_node *pn, *sn; while (1) { if (fn->fn_flags & RTN_TL_ROOT) return NULL; - pn = fn->parent; - if (FIB6_SUBTREE(pn) && FIB6_SUBTREE(pn) != fn) - fn = fib6_lookup(FIB6_SUBTREE(pn), NULL, saddr); + pn = rcu_dereference(fn->parent); + sn = FIB6_SUBTREE(pn); + if (sn && sn != fn) + fn = fib6_lookup(sn, NULL, saddr); else fn = pn; if (fn->fn_flags & RTN_RTINFO) @@ -916,13 +923,19 @@ static struct rt6_info *ip6_pol_route_lookup(struct net *net, struct rt6_info *rt, *rt_cache; struct fib6_node *fn; - read_lock_bh(&table->tb6_lock); + rcu_read_lock(); fn = fib6_lookup(&table->tb6_root, &fl6->daddr, &fl6->saddr); restart: - rt = fn->leaf; - rt = rt6_device_match(net, rt, &fl6->saddr, fl6->flowi6_oif, flags); - if (rt->rt6i_nsiblings && fl6->flowi6_oif == 0) - rt = rt6_multipath_select(rt, fl6, fl6->flowi6_oif, flags); + rt = rcu_dereference(fn->leaf); + if (!rt) { + rt = net->ipv6.ip6_null_entry; + } else { + rt = rt6_device_match(net, rt, &fl6->saddr, + fl6->flowi6_oif, flags); + if (rt->rt6i_nsiblings && fl6->flowi6_oif == 0) + rt = rt6_multipath_select(rt, fl6, + fl6->flowi6_oif, flags); + } if (rt == net->ipv6.ip6_null_entry) { fn = fib6_backtrack(fn, &fl6->saddr); if (fn) @@ -936,7 +949,7 @@ restart: if (ip6_hold_safe(net, &rt, true)) dst_use_noref(&rt->dst, jiffies); - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); trace_fib6_table_lookup(net, rt, table->tb6_id, fl6); @@ -990,9 +1003,9 @@ static int __ip6_ins_rt(struct rt6_info *rt, struct nl_info *info, struct fib6_table *table; table = rt->rt6i_table; - write_lock_bh(&table->tb6_lock); + spin_lock_bh(&table->tb6_lock); err = fib6_add(&table->tb6_root, rt, info, mxc, extack); - write_unlock_bh(&table->tb6_lock); + spin_unlock_bh(&table->tb6_lock); return err; } @@ -1090,7 +1103,7 @@ static struct rt6_info *ip6_rt_pcpu_alloc(struct rt6_info *rt) return pcpu_rt; } -/* It should be called with read_lock_bh(&tb6_lock) acquired */ +/* It should be called with rcu_read_lock() acquired */ static struct rt6_info *rt6_get_pcpu_route(struct rt6_info *rt) { struct rt6_info *pcpu_rt, **p; @@ -1632,7 +1645,7 @@ struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table, if (net->ipv6.devconf_all->forwarding == 0) strict |= RT6_LOOKUP_F_REACHABLE; - read_lock_bh(&table->tb6_lock); + rcu_read_lock(); fn = fib6_lookup(&table->tb6_root, &fl6->daddr, &fl6->saddr); saved_fn = fn; @@ -1662,7 +1675,7 @@ redo_rt6_select: rt = rt_cache; if (rt == net->ipv6.ip6_null_entry) { - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); dst_hold(&rt->dst); trace_fib6_table_lookup(net, rt, table->tb6_id, fl6); return rt; @@ -1671,7 +1684,7 @@ redo_rt6_select: dst_use_noref(&rt->dst, jiffies); rt6_dst_from_metrics_check(rt); } - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); trace_fib6_table_lookup(net, rt, table->tb6_id, fl6); return rt; } else if (unlikely((fl6->flowi6_flags & FLOWI_FLAG_KNOWN_NH) && @@ -1687,11 +1700,11 @@ redo_rt6_select: if (ip6_hold_safe(net, &rt, true)) { dst_use_noref(&rt->dst, jiffies); } else { - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); uncached_rt = rt; goto uncached_rt_out; } - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); uncached_rt = ip6_rt_cache_alloc(rt, &fl6->daddr, NULL); dst_release(&rt->dst); @@ -1719,7 +1732,7 @@ uncached_rt_out: pcpu_rt = rt6_get_pcpu_route(rt); if (pcpu_rt) { - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); } else { /* atomic_inc_not_zero() is needed when using rcu */ if (atomic_inc_not_zero(&rt->rt6i_ref)) { @@ -1730,12 +1743,12 @@ uncached_rt_out: * No dst_hold() on rt is needed because grabbing * rt->rt6i_ref makes sure rt can't be released. */ - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); pcpu_rt = rt6_make_pcpu_route(rt); rt6_release(rt); } else { /* rt is already removed from tree */ - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); pcpu_rt = net->ipv6.ip6_null_entry; dst_hold(&pcpu_rt->dst); } @@ -2131,10 +2144,10 @@ static struct rt6_info *__ip6_route_redirect(struct net *net, * routes. */ - read_lock_bh(&table->tb6_lock); + rcu_read_lock(); fn = fib6_lookup(&table->tb6_root, &fl6->daddr, &fl6->saddr); restart: - for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) { + for_each_fib6_node_rt_rcu(fn) { if (rt6_check_expired(rt)) continue; if (rt->dst.error) @@ -2179,7 +2192,7 @@ restart: out: ip6_hold_safe(net, &rt, true); - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); trace_fib6_table_lookup(net, rt, table->tb6_id, fl6); return rt; @@ -2778,9 +2791,9 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info) } table = rt->rt6i_table; - write_lock_bh(&table->tb6_lock); + spin_lock_bh(&table->tb6_lock); err = fib6_del(rt, info); - write_unlock_bh(&table->tb6_lock); + spin_unlock_bh(&table->tb6_lock); out: ip6_rt_put(rt); @@ -2806,7 +2819,7 @@ static int __ip6_del_rt_siblings(struct rt6_info *rt, struct fib6_config *cfg) if (rt == net->ipv6.ip6_null_entry) goto out_put; table = rt->rt6i_table; - write_lock_bh(&table->tb6_lock); + spin_lock_bh(&table->tb6_lock); if (rt->rt6i_nsiblings && cfg->fc_delete_all_nh) { struct rt6_info *sibling, *next_sibling; @@ -2836,7 +2849,7 @@ static int __ip6_del_rt_siblings(struct rt6_info *rt, struct fib6_config *cfg) err = fib6_del(rt, info); out_unlock: - write_unlock_bh(&table->tb6_lock); + spin_unlock_bh(&table->tb6_lock); out_put: ip6_rt_put(rt); @@ -2861,7 +2874,7 @@ static int ip6_route_del(struct fib6_config *cfg, return err; } - read_lock_bh(&table->tb6_lock); + rcu_read_lock(); fn = fib6_locate(&table->tb6_root, &cfg->fc_dst, cfg->fc_dst_len, @@ -2869,7 +2882,7 @@ static int ip6_route_del(struct fib6_config *cfg, !(cfg->fc_flags & RTF_CACHE)); if (fn) { - for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) { + for_each_fib6_node_rt_rcu(fn) { if (cfg->fc_flags & RTF_CACHE) { rt_cache = rt6_find_cached_rt(rt, &cfg->fc_dst, &cfg->fc_src); @@ -2890,7 +2903,7 @@ static int ip6_route_del(struct fib6_config *cfg, continue; if (!dst_hold_safe(&rt->dst)) break; - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); /* if gateway was specified only delete the one hop */ if (cfg->fc_flags & RTF_GATEWAY) @@ -2899,7 +2912,7 @@ static int ip6_route_del(struct fib6_config *cfg, return __ip6_del_rt_siblings(rt, cfg); } } - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); return err; } @@ -3074,12 +3087,12 @@ static struct rt6_info *rt6_get_route_info(struct net *net, if (!table) return NULL; - read_lock_bh(&table->tb6_lock); + rcu_read_lock(); fn = fib6_locate(&table->tb6_root, prefix, prefixlen, NULL, 0, true); if (!fn) goto out; - for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) { + for_each_fib6_node_rt_rcu(fn) { if (rt->dst.dev->ifindex != ifindex) continue; if ((rt->rt6i_flags & (RTF_ROUTEINFO|RTF_GATEWAY)) != (RTF_ROUTEINFO|RTF_GATEWAY)) @@ -3090,7 +3103,7 @@ static struct rt6_info *rt6_get_route_info(struct net *net, break; } out: - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); return rt; } @@ -3136,8 +3149,8 @@ struct rt6_info *rt6_get_dflt_router(const struct in6_addr *addr, struct net_dev if (!table) return NULL; - read_lock_bh(&table->tb6_lock); - for (rt = table->tb6_root.leaf; rt; rt = rt->dst.rt6_next) { + rcu_read_lock(); + for_each_fib6_node_rt_rcu(&table->tb6_root) { if (dev == rt->dst.dev && ((rt->rt6i_flags & (RTF_ADDRCONF | RTF_DEFAULT)) == (RTF_ADDRCONF | RTF_DEFAULT)) && ipv6_addr_equal(&rt->rt6i_gateway, addr)) @@ -3145,7 +3158,7 @@ struct rt6_info *rt6_get_dflt_router(const struct in6_addr *addr, struct net_dev } if (rt) ip6_hold_safe(NULL, &rt, false); - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); return rt; } @@ -3183,20 +3196,20 @@ static void __rt6_purge_dflt_routers(struct fib6_table *table) struct rt6_info *rt; restart: - read_lock_bh(&table->tb6_lock); - for (rt = table->tb6_root.leaf; rt; rt = rt->dst.rt6_next) { + rcu_read_lock(); + for_each_fib6_node_rt_rcu(&table->tb6_root) { if (rt->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF) && (!rt->rt6i_idev || rt->rt6i_idev->cnf.accept_ra != 2)) { if (dst_hold_safe(&rt->dst)) { - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); ip6_del_rt(rt); } else { - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); } goto restart; } } - read_unlock_bh(&table->tb6_lock); + rcu_read_unlock(); table->flags &= ~RT6_TABLE_HAS_DFLT_ROUTER; } -- cgit v1.2.3 From 81eb8447daae3b62247aa66bb17b82f8fef68249 Mon Sep 17 00:00:00 2001 From: Wei Wang Date: Fri, 6 Oct 2017 12:06:11 -0700 Subject: ipv6: take care of rt6_stats Currently, most of the rt6_stats are not hooked up correctly. As the last part of this patch series, hook up all existing rt6_stats and add one new stat fib_rt_uncache to indicate the number of routes in the uncached list. For details of the stats, please refer to the comments added in include/net/ip6_fib.h. Note: fib_rt_alloc and fib_rt_uncache are not guaranteed to be modified under a lock. So atomic_t is used for them. Signed-off-by: Wei Wang Signed-off-by: Martin KaFai Lau Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 15 +++++++++------ net/ipv6/ip6_fib.c | 42 ++++++++++++++++++++++++------------------ net/ipv6/route.c | 16 ++++++++++++++-- 3 files changed, 47 insertions(+), 26 deletions(-) (limited to 'include/net') diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 0b438b9bcb10..10c913816032 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -297,12 +297,15 @@ struct fib6_walker { }; struct rt6_statistics { - __u32 fib_nodes; - __u32 fib_route_nodes; - __u32 fib_rt_alloc; /* permanent routes */ - __u32 fib_rt_entries; /* rt entries in table */ - __u32 fib_rt_cache; /* cache routes */ - __u32 fib_discarded_routes; + __u32 fib_nodes; /* all fib6 nodes */ + __u32 fib_route_nodes; /* intermediate nodes */ + __u32 fib_rt_entries; /* rt entries in fib table */ + __u32 fib_rt_cache; /* cached rt entries in exception table */ + __u32 fib_discarded_routes; /* total number of routes delete */ + + /* The following stats are not protected by any lock */ + atomic_t fib_rt_alloc; /* total number of routes alloced */ + atomic_t fib_rt_uncache; /* rt entries in uncached list */ }; #define RTN_TL_ROOT 0x0001 diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 3f95908b39c3..52a29ba32928 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -149,18 +149,21 @@ static __be32 addr_bit_set(const void *token, int fn_bit) addr[fn_bit >> 5]; } -static struct fib6_node *node_alloc(void) +static struct fib6_node *node_alloc(struct net *net) { struct fib6_node *fn; fn = kmem_cache_zalloc(fib6_node_kmem, GFP_ATOMIC); + if (fn) + net->ipv6.rt6_stats->fib_nodes++; return fn; } -static void node_free_immediate(struct fib6_node *fn) +static void node_free_immediate(struct net *net, struct fib6_node *fn) { kmem_cache_free(fib6_node_kmem, fn); + net->ipv6.rt6_stats->fib_nodes--; } static void node_free_rcu(struct rcu_head *head) @@ -170,9 +173,10 @@ static void node_free_rcu(struct rcu_head *head) kmem_cache_free(fib6_node_kmem, fn); } -static void node_free(struct fib6_node *fn) +static void node_free(struct net *net, struct fib6_node *fn) { call_rcu(&fn->rcu, node_free_rcu); + net->ipv6.rt6_stats->fib_nodes--; } void rt6_free_pcpu(struct rt6_info *non_pcpu_rt) @@ -583,7 +587,8 @@ out: * node. */ -static struct fib6_node *fib6_add_1(struct fib6_table *table, +static struct fib6_node *fib6_add_1(struct net *net, + struct fib6_table *table, struct fib6_node *root, struct in6_addr *addr, int plen, int offset, int allow_create, @@ -675,7 +680,7 @@ static struct fib6_node *fib6_add_1(struct fib6_table *table, * Create new leaf node without children. */ - ln = node_alloc(); + ln = node_alloc(net); if (!ln) return ERR_PTR(-ENOMEM); @@ -716,14 +721,14 @@ insert_above: * (new leaf node)[ln] (old node)[fn] */ if (plen > bit) { - in = node_alloc(); - ln = node_alloc(); + in = node_alloc(net); + ln = node_alloc(net); if (!in || !ln) { if (in) - node_free_immediate(in); + node_free_immediate(net, in); if (ln) - node_free_immediate(ln); + node_free_immediate(net, ln); return ERR_PTR(-ENOMEM); } @@ -768,7 +773,7 @@ insert_above: * (old node)[fn] NULL */ - ln = node_alloc(); + ln = node_alloc(net); if (!ln) return ERR_PTR(-ENOMEM); @@ -1065,6 +1070,7 @@ add: fn->rr_ptr = NULL; rt6_release(iter); nsiblings--; + info->nl_net->ipv6.rt6_stats->fib_rt_entries--; } else { ins = &iter->dst.rt6_next; } @@ -1140,7 +1146,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, if (!allow_create && !replace_required) pr_warn("RTM_NEWROUTE with no NLM_F_CREATE or NLM_F_REPLACE\n"); - fn = fib6_add_1(table, root, + fn = fib6_add_1(info->nl_net, table, root, &rt->rt6i_dst.addr, rt->rt6i_dst.plen, offsetof(struct rt6_info, rt6i_dst), allow_create, replace_required, extack); @@ -1170,7 +1176,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, */ /* Create subtree root node */ - sfn = node_alloc(); + sfn = node_alloc(info->nl_net); if (!sfn) goto failure; @@ -1181,8 +1187,8 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, /* Now add the first leaf node to new subtree */ - sn = fib6_add_1(table, sfn, &rt->rt6i_src.addr, - rt->rt6i_src.plen, + sn = fib6_add_1(info->nl_net, table, sfn, + &rt->rt6i_src.addr, rt->rt6i_src.plen, offsetof(struct rt6_info, rt6i_src), allow_create, replace_required, extack); @@ -1191,7 +1197,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, root, and then (in failure) stale node in main tree. */ - node_free_immediate(sfn); + node_free_immediate(info->nl_net, sfn); err = PTR_ERR(sn); goto failure; } @@ -1200,8 +1206,8 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, rcu_assign_pointer(sfn->parent, fn); rcu_assign_pointer(fn->subtree, sfn); } else { - sn = fib6_add_1(table, FIB6_SUBTREE(fn), &rt->rt6i_src.addr, - rt->rt6i_src.plen, + sn = fib6_add_1(info->nl_net, table, FIB6_SUBTREE(fn), + &rt->rt6i_src.addr, rt->rt6i_src.plen, offsetof(struct rt6_info, rt6i_src), allow_create, replace_required, extack); @@ -1609,7 +1615,7 @@ static struct fib6_node *fib6_repair_tree(struct net *net, } read_unlock(&net->ipv6.fib6_walker_lock); - node_free(fn); + node_free(net, fn); if (pn->fn_flags & RTN_RTINFO || FIB6_SUBTREE(pn)) return pn; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index cf44d0994b1e..399d1bceec4a 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -143,9 +143,11 @@ static void rt6_uncached_list_del(struct rt6_info *rt) { if (!list_empty(&rt->rt6i_uncached)) { struct uncached_list *ul = rt->rt6i_uncached_list; + struct net *net = dev_net(rt->dst.dev); spin_lock_bh(&ul->lock); list_del(&rt->rt6i_uncached); + atomic_dec(&net->ipv6.rt6_stats->fib_rt_uncache); spin_unlock_bh(&ul->lock); } } @@ -359,8 +361,10 @@ static struct rt6_info *__ip6_dst_alloc(struct net *net, struct rt6_info *rt = dst_alloc(&net->ipv6.ip6_dst_ops, dev, 1, DST_OBSOLETE_FORCE_CHK, flags); - if (rt) + if (rt) { rt6_info_init(rt); + atomic_inc(&net->ipv6.rt6_stats->fib_rt_alloc); + } return rt; } @@ -1156,6 +1160,8 @@ static DEFINE_SPINLOCK(rt6_exception_lock); static void rt6_remove_exception(struct rt6_exception_bucket *bucket, struct rt6_exception *rt6_ex) { + struct net *net = dev_net(rt6_ex->rt6i->dst.dev); + if (!bucket || !rt6_ex) return; rt6_ex->rt6i->rt6i_node = NULL; @@ -1164,6 +1170,7 @@ static void rt6_remove_exception(struct rt6_exception_bucket *bucket, kfree_rcu(rt6_ex, rcu); WARN_ON_ONCE(!bucket->depth); bucket->depth--; + net->ipv6.rt6_stats->fib_rt_cache--; } /* Remove oldest rt6_ex in bucket and free the memory @@ -1270,6 +1277,7 @@ __rt6_find_exception_rcu(struct rt6_exception_bucket **bucket, static int rt6_insert_exception(struct rt6_info *nrt, struct rt6_info *ort) { + struct net *net = dev_net(ort->dst.dev); struct rt6_exception_bucket *bucket; struct in6_addr *src_key = NULL; struct rt6_exception *rt6_ex; @@ -1339,6 +1347,7 @@ static int rt6_insert_exception(struct rt6_info *nrt, nrt->rt6i_node = ort->rt6i_node; hlist_add_head_rcu(&rt6_ex->hlist, &bucket->chain); bucket->depth++; + net->ipv6.rt6_stats->fib_rt_cache++; if (bucket->depth > FIB6_MAX_DEPTH) rt6_exception_remove_oldest(bucket); @@ -1714,6 +1723,7 @@ redo_rt6_select: * No need for another dst_hold() */ rt6_uncached_list_add(uncached_rt); + atomic_inc(&net->ipv6.rt6_stats->fib_rt_uncache); } else { uncached_rt = net->ipv6.ip6_null_entry; dst_hold(&uncached_rt->dst); @@ -1894,6 +1904,7 @@ struct dst_entry *ip6_blackhole_route(struct net *net, struct dst_entry *dst_ori DST_OBSOLETE_NONE, 0); if (rt) { rt6_info_init(rt); + atomic_inc(&net->ipv6.rt6_stats->fib_rt_alloc); new = &rt->dst; new->__use = 1; @@ -2341,6 +2352,7 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev, * do proper release of the net_device */ rt6_uncached_list_add(rt); + atomic_inc(&net->ipv6.rt6_stats->fib_rt_uncache); dst = xfrm_lookup(net, &rt->dst, flowi6_to_flowi(fl6), NULL, 0); @@ -4422,7 +4434,7 @@ static int rt6_stats_seq_show(struct seq_file *seq, void *v) seq_printf(seq, "%04x %04x %04x %04x %04x %04x %04x\n", net->ipv6.rt6_stats->fib_nodes, net->ipv6.rt6_stats->fib_route_nodes, - net->ipv6.rt6_stats->fib_rt_alloc, + atomic_read(&net->ipv6.rt6_stats->fib_rt_alloc), net->ipv6.rt6_stats->fib_rt_entries, net->ipv6.rt6_stats->fib_rt_cache, dst_entries_get_slow(&net->ipv6.ip6_dst_ops), -- cgit v1.2.3 From 548ec114705bb8f0879a0da12abec17f17a7cc26 Mon Sep 17 00:00:00 2001 From: Lin Zhang Date: Fri, 6 Oct 2017 01:40:35 +0800 Subject: net: phonet: mark phonet_protocol as const The phonet_protocol structs don't need to be written by anyone and so can be marked as const. Signed-off-by: Lin Zhang Signed-off-by: David S. Miller --- include/net/phonet/phonet.h | 6 ++++-- net/phonet/af_phonet.c | 15 ++++++++------- net/phonet/datagram.c | 2 +- net/phonet/pep.c | 2 +- 4 files changed, 14 insertions(+), 11 deletions(-) (limited to 'include/net') diff --git a/include/net/phonet/phonet.h b/include/net/phonet/phonet.h index 039cc29cb4a8..51e1a2a45d02 100644 --- a/include/net/phonet/phonet.h +++ b/include/net/phonet/phonet.h @@ -108,8 +108,10 @@ struct phonet_protocol { int sock_type; }; -int phonet_proto_register(unsigned int protocol, struct phonet_protocol *pp); -void phonet_proto_unregister(unsigned int protocol, struct phonet_protocol *pp); +int phonet_proto_register(unsigned int protocol, + const struct phonet_protocol *pp); +void phonet_proto_unregister(unsigned int protocol, + const struct phonet_protocol *pp); int phonet_sysctl_init(void); void phonet_sysctl_exit(void); diff --git a/net/phonet/af_phonet.c b/net/phonet/af_phonet.c index b12142e55d19..3b0ef691f5b1 100644 --- a/net/phonet/af_phonet.c +++ b/net/phonet/af_phonet.c @@ -35,11 +35,11 @@ #include /* Transport protocol registration */ -static struct phonet_protocol *proto_tab[PHONET_NPROTO] __read_mostly; +static const struct phonet_protocol *proto_tab[PHONET_NPROTO] __read_mostly; -static struct phonet_protocol *phonet_proto_get(unsigned int protocol) +static const struct phonet_protocol *phonet_proto_get(unsigned int protocol) { - struct phonet_protocol *pp; + const struct phonet_protocol *pp; if (protocol >= PHONET_NPROTO) return NULL; @@ -53,7 +53,7 @@ static struct phonet_protocol *phonet_proto_get(unsigned int protocol) return pp; } -static inline void phonet_proto_put(struct phonet_protocol *pp) +static inline void phonet_proto_put(const struct phonet_protocol *pp) { module_put(pp->prot->owner); } @@ -65,7 +65,7 @@ static int pn_socket_create(struct net *net, struct socket *sock, int protocol, { struct sock *sk; struct pn_sock *pn; - struct phonet_protocol *pnp; + const struct phonet_protocol *pnp; int err; if (!capable(CAP_SYS_ADMIN)) @@ -470,7 +470,7 @@ static struct packet_type phonet_packet_type __read_mostly = { static DEFINE_MUTEX(proto_tab_lock); int __init_or_module phonet_proto_register(unsigned int protocol, - struct phonet_protocol *pp) + const struct phonet_protocol *pp) { int err = 0; @@ -492,7 +492,8 @@ int __init_or_module phonet_proto_register(unsigned int protocol, } EXPORT_SYMBOL(phonet_proto_register); -void phonet_proto_unregister(unsigned int protocol, struct phonet_protocol *pp) +void phonet_proto_unregister(unsigned int protocol, + const struct phonet_protocol *pp) { mutex_lock(&proto_tab_lock); BUG_ON(proto_tab[protocol] != pp); diff --git a/net/phonet/datagram.c b/net/phonet/datagram.c index 5e710435ffa9..b44fb9018fb8 100644 --- a/net/phonet/datagram.c +++ b/net/phonet/datagram.c @@ -195,7 +195,7 @@ static struct proto pn_proto = { .name = "PHONET", }; -static struct phonet_protocol pn_dgram_proto = { +static const struct phonet_protocol pn_dgram_proto = { .ops = &phonet_dgram_ops, .prot = &pn_proto, .sock_type = SOCK_DGRAM, diff --git a/net/phonet/pep.c b/net/phonet/pep.c index e81537991ddf..9fc76b19cd3c 100644 --- a/net/phonet/pep.c +++ b/net/phonet/pep.c @@ -1351,7 +1351,7 @@ static struct proto pep_proto = { .name = "PNPIPE", }; -static struct phonet_protocol pep_pn_proto = { +static const struct phonet_protocol pep_pn_proto = { .ops = &phonet_stream_ops, .prot = &pep_proto, .sock_type = SOCK_SEQPACKET, -- cgit v1.2.3 From 77041420751fe6d4acf2103b245dcc2b4b7b8360 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Mon, 9 Oct 2017 11:15:31 +0200 Subject: net: bridge: Notify on bridge device mrouter state changes Add the SWITCHDEV_ATTR_ID_BRIDGE_MROUTER switchdev notification type, used to indicate whether the bridge is or isn't mrouter. Notify when the bridge changes its state, similarly to the already existing bridged port mrouter notifications. The notification uses the switchdev_attr.u.mrouter boolean flag to indicate the current bridge mrouter status. Thus, it only indicates whether the bridge is currently used as an mrouter or not, and does not indicate the exact mrouter state of the bridge (learning, permanent, etc.). Signed-off-by: Yotam Gigi Signed-off-by: Jiri Pirko Signed-off-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/net/switchdev.h | 1 + net/bridge/br_multicast.c | 38 +++++++++++++++++++++++++++++++++++--- 2 files changed, 36 insertions(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/switchdev.h b/include/net/switchdev.h index d767b7991887..d756fbe46625 100644 --- a/include/net/switchdev.h +++ b/include/net/switchdev.h @@ -51,6 +51,7 @@ enum switchdev_attr_id { SWITCHDEV_ATTR_ID_BRIDGE_AGEING_TIME, SWITCHDEV_ATTR_ID_BRIDGE_VLAN_FILTERING, SWITCHDEV_ATTR_ID_BRIDGE_MC_DISABLED, + SWITCHDEV_ATTR_ID_BRIDGE_MROUTER, }; struct switchdev_attr { diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 8dc5c8d69bcd..bd50550dd4ca 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -859,8 +859,32 @@ out: spin_unlock(&br->multicast_lock); } +static void br_mc_router_state_change(struct net_bridge *p, + bool is_mc_router) +{ + struct switchdev_attr attr = { + .orig_dev = p->dev, + .id = SWITCHDEV_ATTR_ID_BRIDGE_MROUTER, + .flags = SWITCHDEV_F_DEFER, + .u.mrouter = is_mc_router, + }; + + switchdev_port_attr_set(p->dev, &attr); +} + static void br_multicast_local_router_expired(unsigned long data) { + struct net_bridge *br = (struct net_bridge *)data; + + spin_lock(&br->multicast_lock); + if (br->multicast_router == MDB_RTR_TYPE_DISABLED || + br->multicast_router == MDB_RTR_TYPE_PERM || + timer_pending(&br->multicast_router_timer)) + goto out; + + br_mc_router_state_change(br, false); +out: + spin_unlock(&br->multicast_lock); } static void br_multicast_querier_expired(struct net_bridge *br, @@ -1364,9 +1388,12 @@ static void br_multicast_mark_router(struct net_bridge *br, unsigned long now = jiffies; if (!port) { - if (br->multicast_router == MDB_RTR_TYPE_TEMP_QUERY) + if (br->multicast_router == MDB_RTR_TYPE_TEMP_QUERY) { + if (!timer_pending(&br->multicast_router_timer)) + br_mc_router_state_change(br, true); mod_timer(&br->multicast_router_timer, now + br->multicast_querier_interval); + } return; } @@ -1952,7 +1979,7 @@ void br_multicast_init(struct net_bridge *br) spin_lock_init(&br->multicast_lock); setup_timer(&br->multicast_router_timer, - br_multicast_local_router_expired, 0); + br_multicast_local_router_expired, (unsigned long)br); setup_timer(&br->ip4_other_query.timer, br_ip4_multicast_querier_expired, (unsigned long)br); setup_timer(&br->ip4_own_query.timer, br_ip4_multicast_query_expired, @@ -2042,9 +2069,14 @@ int br_multicast_set_router(struct net_bridge *br, unsigned long val) switch (val) { case MDB_RTR_TYPE_DISABLED: case MDB_RTR_TYPE_PERM: + br_mc_router_state_change(br, val == MDB_RTR_TYPE_PERM); del_timer(&br->multicast_router_timer); - /* fall through */ + br->multicast_router = val; + err = 0; + break; case MDB_RTR_TYPE_TEMP_QUERY: + if (br->multicast_router != MDB_RTR_TYPE_TEMP_QUERY) + br_mc_router_state_change(br, false); br->multicast_router = val; err = 0; break; -- cgit v1.2.3 From d66f2b91f95b56e31772b9faa0d036cd2e53cb02 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 9 Oct 2017 10:30:14 -0700 Subject: bpf: don't rely on the verifier lock for metadata_dst allocation bpf_skb_set_tunnel_*() functions require allocation of per-cpu metadata_dst. The allocation happens upon verification of the first program using those helpers. In preparation for removing the verifier lock, use cmpxchg() to make sure we only allocate the metadata_dsts once. Signed-off-by: Jakub Kicinski Reviewed-by: Simon Horman Acked-by: Alexei Starovoitov Acked-by: Daniel Borkmann Signed-off-by: David S. Miller --- include/net/dst_metadata.h | 1 + net/core/dst.c | 16 ++++++++++++++++ net/core/filter.c | 16 +++++++++------- 3 files changed, 26 insertions(+), 7 deletions(-) (limited to 'include/net') diff --git a/include/net/dst_metadata.h b/include/net/dst_metadata.h index 9fba2ebf6dda..87a0bb8d449f 100644 --- a/include/net/dst_metadata.h +++ b/include/net/dst_metadata.h @@ -87,6 +87,7 @@ static inline int skb_metadata_dst_cmp(const struct sk_buff *skb_a, void metadata_dst_free(struct metadata_dst *); struct metadata_dst *metadata_dst_alloc(u8 optslen, enum metadata_type type, gfp_t flags); +void metadata_dst_free_percpu(struct metadata_dst __percpu *md_dst); struct metadata_dst __percpu * metadata_dst_alloc_percpu(u8 optslen, enum metadata_type type, gfp_t flags); diff --git a/net/core/dst.c b/net/core/dst.c index a6c47da7d0f8..8b2eafac984d 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -322,3 +322,19 @@ metadata_dst_alloc_percpu(u8 optslen, enum metadata_type type, gfp_t flags) return md_dst; } EXPORT_SYMBOL_GPL(metadata_dst_alloc_percpu); + +void metadata_dst_free_percpu(struct metadata_dst __percpu *md_dst) +{ + int cpu; + +#ifdef CONFIG_DST_CACHE + for_each_possible_cpu(cpu) { + struct metadata_dst *one_md_dst = per_cpu_ptr(md_dst, cpu); + + if (one_md_dst->type == METADATA_IP_TUNNEL) + dst_cache_destroy(&one_md_dst->u.tun_info.dst_cache); + } +#endif + free_percpu(md_dst); +} +EXPORT_SYMBOL_GPL(metadata_dst_free_percpu); diff --git a/net/core/filter.c b/net/core/filter.c index b7e8caa1e790..140fa9f9c0f4 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -43,6 +43,7 @@ #include #include #include +#include #include #include #include @@ -2987,14 +2988,15 @@ static const struct bpf_func_proto * bpf_get_skb_set_tunnel_proto(enum bpf_func_id which) { if (!md_dst) { - /* Race is not possible, since it's called from verifier - * that is holding verifier mutex. - */ - md_dst = metadata_dst_alloc_percpu(IP_TUNNEL_OPTS_MAX, - METADATA_IP_TUNNEL, - GFP_KERNEL); - if (!md_dst) + struct metadata_dst __percpu *tmp; + + tmp = metadata_dst_alloc_percpu(IP_TUNNEL_OPTS_MAX, + METADATA_IP_TUNNEL, + GFP_KERNEL); + if (!tmp) return NULL; + if (cmpxchg(&md_dst, NULL, tmp)) + metadata_dst_free_percpu(tmp); } switch (which) { -- cgit v1.2.3 From 8c418b5b15747eda05d086e80fa0a767982fbf37 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Fri, 6 Oct 2017 11:53:32 +0200 Subject: fq: support filtering a given tin MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add to the FQ API a way to filter a given tin, in order to remove frames that fulfil certain criteria according to a filter function. This will be used by mac80211 to remove frames belonging to an AP VLAN interface that's being removed. Signed-off-by: Johannes Berg Acked-by: Toke Høiland-Jørgensen Signed-off-by: Johannes Berg --- include/net/fq.h | 7 +++++ include/net/fq_impl.h | 72 ++++++++++++++++++++++++++++++++++++++++++++------- 2 files changed, 69 insertions(+), 10 deletions(-) (limited to 'include/net') diff --git a/include/net/fq.h b/include/net/fq.h index 6d8521a30c5c..ac944a686840 100644 --- a/include/net/fq.h +++ b/include/net/fq.h @@ -90,6 +90,13 @@ typedef void fq_skb_free_t(struct fq *, struct fq_flow *, struct sk_buff *); +/* Return %true to filter (drop) the frame. */ +typedef bool fq_skb_filter_t(struct fq *, + struct fq_tin *, + struct fq_flow *, + struct sk_buff *, + void *); + typedef struct fq_flow *fq_flow_get_default_t(struct fq *, struct fq_tin *, int idx, diff --git a/include/net/fq_impl.h b/include/net/fq_impl.h index 4e6131cd3f43..8b237e4afee6 100644 --- a/include/net/fq_impl.h +++ b/include/net/fq_impl.h @@ -12,24 +12,22 @@ /* functions that are embedded into includer */ -static struct sk_buff *fq_flow_dequeue(struct fq *fq, - struct fq_flow *flow) +static void fq_adjust_removal(struct fq *fq, + struct fq_flow *flow, + struct sk_buff *skb) { struct fq_tin *tin = flow->tin; - struct fq_flow *i; - struct sk_buff *skb; - - lockdep_assert_held(&fq->lock); - - skb = __skb_dequeue(&flow->queue); - if (!skb) - return NULL; tin->backlog_bytes -= skb->len; tin->backlog_packets--; flow->backlog -= skb->len; fq->backlog--; fq->memory_usage -= skb->truesize; +} + +static void fq_rejigger_backlog(struct fq *fq, struct fq_flow *flow) +{ + struct fq_flow *i; if (flow->backlog == 0) { list_del_init(&flow->backlogchain); @@ -43,6 +41,21 @@ static struct sk_buff *fq_flow_dequeue(struct fq *fq, list_move_tail(&flow->backlogchain, &i->backlogchain); } +} + +static struct sk_buff *fq_flow_dequeue(struct fq *fq, + struct fq_flow *flow) +{ + struct sk_buff *skb; + + lockdep_assert_held(&fq->lock); + + skb = __skb_dequeue(&flow->queue); + if (!skb) + return NULL; + + fq_adjust_removal(fq, flow, skb); + fq_rejigger_backlog(fq, flow); return skb; } @@ -188,6 +201,45 @@ static void fq_tin_enqueue(struct fq *fq, } } +static void fq_flow_filter(struct fq *fq, + struct fq_flow *flow, + fq_skb_filter_t filter_func, + void *filter_data, + fq_skb_free_t free_func) +{ + struct fq_tin *tin = flow->tin; + struct sk_buff *skb, *tmp; + + lockdep_assert_held(&fq->lock); + + skb_queue_walk_safe(&flow->queue, skb, tmp) { + if (!filter_func(fq, tin, flow, skb, filter_data)) + continue; + + __skb_unlink(skb, &flow->queue); + fq_adjust_removal(fq, flow, skb); + free_func(fq, tin, flow, skb); + } + + fq_rejigger_backlog(fq, flow); +} + +static void fq_tin_filter(struct fq *fq, + struct fq_tin *tin, + fq_skb_filter_t filter_func, + void *filter_data, + fq_skb_free_t free_func) +{ + struct fq_flow *flow; + + lockdep_assert_held(&fq->lock); + + list_for_each_entry(flow, &tin->new_flows, flowchain) + fq_flow_filter(fq, flow, filter_func, filter_data, free_func); + list_for_each_entry(flow, &tin->old_flows, flowchain) + fq_flow_filter(fq, flow, filter_func, filter_data, free_func); +} + static void fq_flow_reset(struct fq *fq, struct fq_flow *flow, fq_skb_free_t free_func) -- cgit v1.2.3 From 4a269818a7eb8577d32d8b2879099c689ddbd856 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 11 Oct 2017 13:27:29 -0700 Subject: tcp: fix tcp_unlink_write_queue() Yury reported crash with this signature : [ 554.034021] [] 0xffff80003ccd5a58 [ 554.034156] [] skb_release_all+0x14/0x30 [ 554.034288] [] __kfree_skb+0x14/0x28 [ 554.034409] [] tcp_sendmsg_locked+0x4dc/0xcc8 [ 554.034541] [] tcp_sendmsg+0x34/0x58 [ 554.034659] [] inet_sendmsg+0x2c/0xf8 [ 554.034783] [] sock_sendmsg+0x18/0x30 [ 554.034928] [] SyS_sendto+0x84/0xf8 Problem is that skb->destructor contains garbage, and this is because I accidentally removed tcp_skb_tsorted_anchor_cleanup() from tcp_unlink_write_queue() This would trigger with a write(fd, , len) attempt, and we will add to packetdrill this capability to avoid future regressions. Fixes: 75c119afe14f ("tcp: implement rb-tree based retransmit queue") Reported-by: Yury Norov Tested-by: Yury Norov Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/tcp.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/net') diff --git a/include/net/tcp.h b/include/net/tcp.h index 5a95e5886b55..15163454174b 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1712,6 +1712,7 @@ static inline void tcp_insert_write_queue_before(struct sk_buff *new, static inline void tcp_unlink_write_queue(struct sk_buff *skb, struct sock *sk) { + tcp_skb_tsorted_anchor_cleanup(skb); __skb_unlink(skb, &sk->sk_write_queue); } -- cgit v1.2.3 From 843e79d05addd8eb06992cd6dfafc7b9d53f2bc8 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Wed, 11 Oct 2017 09:41:07 +0200 Subject: net: sched: make tc_action_ops->get_dev return dev and avoid passing net Return dev directly, NULL if not possible. That is enough. Makes no sense to pass struct net * to get_dev op, as there is only one net possible, the one the action was created in. So just store it in mirred priv and use directly. Rename the mirred op callback function. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/act_api.h | 3 +-- include/net/tc_act/tc_mirred.h | 1 + net/sched/act_mirred.c | 13 +++++-------- net/sched/cls_api.c | 6 ++---- 4 files changed, 9 insertions(+), 14 deletions(-) (limited to 'include/net') diff --git a/include/net/act_api.h b/include/net/act_api.h index b944e0eb93be..900168a9901e 100644 --- a/include/net/act_api.h +++ b/include/net/act_api.h @@ -93,8 +93,7 @@ struct tc_action_ops { int (*walk)(struct net *, struct sk_buff *, struct netlink_callback *, int, const struct tc_action_ops *); void (*stats_update)(struct tc_action *, u64, u32, u64); - int (*get_dev)(const struct tc_action *a, struct net *net, - struct net_device **mirred_dev); + struct net_device *(*get_dev)(const struct tc_action *a); }; struct tc_action_net { diff --git a/include/net/tc_act/tc_mirred.h b/include/net/tc_act/tc_mirred.h index 604bc31e23ab..21a656569840 100644 --- a/include/net/tc_act/tc_mirred.h +++ b/include/net/tc_act/tc_mirred.h @@ -10,6 +10,7 @@ struct tcf_mirred { int tcfm_ifindex; bool tcfm_mac_header_xmit; struct net_device __rcu *tcfm_dev; + struct net *net; struct list_head tcfm_list; }; #define to_mirred(a) ((struct tcf_mirred *)a) diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index 416627c66f08..8b3e59388480 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -140,6 +140,7 @@ static int tcf_mirred_init(struct net *net, struct nlattr *nla, m->tcfm_eaction = parm->eaction; if (dev != NULL) { m->tcfm_ifindex = parm->ifindex; + m->net = net; if (ret != ACT_P_CREATED) dev_put(rcu_dereference_protected(m->tcfm_dev, 1)); dev_hold(dev); @@ -313,15 +314,11 @@ static struct notifier_block mirred_device_notifier = { .notifier_call = mirred_device_event, }; -static int tcf_mirred_device(const struct tc_action *a, struct net *net, - struct net_device **mirred_dev) +static struct net_device *tcf_mirred_get_dev(const struct tc_action *a) { - int ifindex = tcf_mirred_ifindex(a); + struct tcf_mirred *m = to_mirred(a); - *mirred_dev = __dev_get_by_index(net, ifindex); - if (!*mirred_dev) - return -EINVAL; - return 0; + return __dev_get_by_index(m->net, m->tcfm_ifindex); } static struct tc_action_ops act_mirred_ops = { @@ -336,7 +333,7 @@ static struct tc_action_ops act_mirred_ops = { .walk = tcf_mirred_walker, .lookup = tcf_mirred_search, .size = sizeof(struct tcf_mirred), - .get_dev = tcf_mirred_device, + .get_dev = tcf_mirred_get_dev, }; static __net_init int mirred_init_net(struct net *net) diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 0b2219adf520..450873b0c4b9 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -1016,10 +1016,8 @@ int tcf_exts_get_dev(struct net_device *dev, struct tcf_exts *exts, tcf_exts_to_list(exts, &actions); list_for_each_entry(a, &actions, list) { - if (a->ops->get_dev) { - a->ops->get_dev(a, dev_net(dev), hw_dev); - break; - } + if (a->ops->get_dev) + *hw_dev = a->ops->get_dev(a); } if (*hw_dev) return 0; -- cgit v1.2.3 From b3f55bdda8df55a563005e00b1b71212d8546541 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Wed, 11 Oct 2017 09:41:08 +0200 Subject: net: sched: introduce per-egress action device callbacks Introduce infrastructure that allows drivers to register callbacks that are called whenever tc would offload inserted rule and specified device acts as tc action egress device. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/act_api.h | 34 ++++++++ include/net/pkt_cls.h | 2 + net/sched/act_api.c | 220 ++++++++++++++++++++++++++++++++++++++++++++++++++ net/sched/cls_api.c | 30 +++++++ 4 files changed, 286 insertions(+) (limited to 'include/net') diff --git a/include/net/act_api.h b/include/net/act_api.h index 900168a9901e..f5e8c9048fb0 100644 --- a/include/net/act_api.h +++ b/include/net/act_api.h @@ -174,4 +174,38 @@ static inline void tcf_action_stats_update(struct tc_action *a, u64 bytes, #endif } +typedef int tc_setup_cb_t(enum tc_setup_type type, + void *type_data, void *cb_priv); + +#ifdef CONFIG_NET_CLS_ACT +int tc_setup_cb_egdev_register(const struct net_device *dev, + tc_setup_cb_t *cb, void *cb_priv); +void tc_setup_cb_egdev_unregister(const struct net_device *dev, + tc_setup_cb_t *cb, void *cb_priv); +int tc_setup_cb_egdev_call(const struct net_device *dev, + enum tc_setup_type type, void *type_data, + bool err_stop); +#else +static inline +int tc_setup_cb_egdev_register(const struct net_device *dev, + tc_setup_cb_t *cb, void *cb_priv) +{ + return 0; +} + +static inline +void tc_setup_cb_egdev_unregister(const struct net_device *dev, + tc_setup_cb_t *cb, void *cb_priv) +{ +} + +static inline +int tc_setup_cb_egdev_call(const struct net_device *dev, + enum tc_setup_type type, void *type_data, + bool err_stop) +{ + return 0; +} +#endif + #endif diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index e80edd8879ef..6f8149c82571 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -206,6 +206,8 @@ int tcf_exts_dump(struct sk_buff *skb, struct tcf_exts *exts); int tcf_exts_dump_stats(struct sk_buff *skb, struct tcf_exts *exts); int tcf_exts_get_dev(struct net_device *dev, struct tcf_exts *exts, struct net_device **hw_dev); +int tcf_exts_egdev_cb_call(struct tcf_exts *exts, enum tc_setup_type type, + void *type_data, bool err_stop); /** * struct tcf_pkt_info - packet information diff --git a/net/sched/act_api.c b/net/sched/act_api.c index da6fa82c98a8..ac97db92ab68 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -21,6 +21,8 @@ #include #include #include +#include +#include #include #include #include @@ -1249,8 +1251,226 @@ out_module_put: return skb->len; } +struct tcf_action_net { + struct rhashtable egdev_ht; +}; + +static unsigned int tcf_action_net_id; + +struct tcf_action_egdev_cb { + struct list_head list; + tc_setup_cb_t *cb; + void *cb_priv; +}; + +struct tcf_action_egdev { + struct rhash_head ht_node; + const struct net_device *dev; + unsigned int refcnt; + struct list_head cb_list; +}; + +static const struct rhashtable_params tcf_action_egdev_ht_params = { + .key_offset = offsetof(struct tcf_action_egdev, dev), + .head_offset = offsetof(struct tcf_action_egdev, ht_node), + .key_len = sizeof(const struct net_device *), +}; + +static struct tcf_action_egdev * +tcf_action_egdev_lookup(const struct net_device *dev) +{ + struct net *net = dev_net(dev); + struct tcf_action_net *tan = net_generic(net, tcf_action_net_id); + + return rhashtable_lookup_fast(&tan->egdev_ht, &dev, + tcf_action_egdev_ht_params); +} + +static struct tcf_action_egdev * +tcf_action_egdev_get(const struct net_device *dev) +{ + struct tcf_action_egdev *egdev; + struct tcf_action_net *tan; + + egdev = tcf_action_egdev_lookup(dev); + if (egdev) + goto inc_ref; + + egdev = kzalloc(sizeof(*egdev), GFP_KERNEL); + if (!egdev) + return NULL; + INIT_LIST_HEAD(&egdev->cb_list); + tan = net_generic(dev_net(dev), tcf_action_net_id); + rhashtable_insert_fast(&tan->egdev_ht, &egdev->ht_node, + tcf_action_egdev_ht_params); + +inc_ref: + egdev->refcnt++; + return egdev; +} + +static void tcf_action_egdev_put(struct tcf_action_egdev *egdev) +{ + struct tcf_action_net *tan; + + if (--egdev->refcnt) + return; + tan = net_generic(dev_net(egdev->dev), tcf_action_net_id); + rhashtable_remove_fast(&tan->egdev_ht, &egdev->ht_node, + tcf_action_egdev_ht_params); + kfree(egdev); +} + +static struct tcf_action_egdev_cb * +tcf_action_egdev_cb_lookup(struct tcf_action_egdev *egdev, + tc_setup_cb_t *cb, void *cb_priv) +{ + struct tcf_action_egdev_cb *egdev_cb; + + list_for_each_entry(egdev_cb, &egdev->cb_list, list) + if (egdev_cb->cb == cb && egdev_cb->cb_priv == cb_priv) + return egdev_cb; + return NULL; +} + +static int tcf_action_egdev_cb_call(struct tcf_action_egdev *egdev, + enum tc_setup_type type, + void *type_data, bool err_stop) +{ + struct tcf_action_egdev_cb *egdev_cb; + int ok_count = 0; + int err; + + list_for_each_entry(egdev_cb, &egdev->cb_list, list) { + err = egdev_cb->cb(type, type_data, egdev_cb->cb_priv); + if (err) { + if (err_stop) + return err; + } else { + ok_count++; + } + } + return ok_count; +} + +static int tcf_action_egdev_cb_add(struct tcf_action_egdev *egdev, + tc_setup_cb_t *cb, void *cb_priv) +{ + struct tcf_action_egdev_cb *egdev_cb; + + egdev_cb = tcf_action_egdev_cb_lookup(egdev, cb, cb_priv); + if (WARN_ON(egdev_cb)) + return -EEXIST; + egdev_cb = kzalloc(sizeof(*egdev_cb), GFP_KERNEL); + if (!egdev_cb) + return -ENOMEM; + egdev_cb->cb = cb; + egdev_cb->cb_priv = cb_priv; + list_add(&egdev_cb->list, &egdev->cb_list); + return 0; +} + +static void tcf_action_egdev_cb_del(struct tcf_action_egdev *egdev, + tc_setup_cb_t *cb, void *cb_priv) +{ + struct tcf_action_egdev_cb *egdev_cb; + + egdev_cb = tcf_action_egdev_cb_lookup(egdev, cb, cb_priv); + if (WARN_ON(!egdev_cb)) + return; + list_del(&egdev_cb->list); + kfree(egdev_cb); +} + +static int __tc_setup_cb_egdev_register(const struct net_device *dev, + tc_setup_cb_t *cb, void *cb_priv) +{ + struct tcf_action_egdev *egdev = tcf_action_egdev_get(dev); + int err; + + if (!egdev) + return -ENOMEM; + err = tcf_action_egdev_cb_add(egdev, cb, cb_priv); + if (err) + goto err_cb_add; + return 0; + +err_cb_add: + tcf_action_egdev_put(egdev); + return err; +} +int tc_setup_cb_egdev_register(const struct net_device *dev, + tc_setup_cb_t *cb, void *cb_priv) +{ + int err; + + rtnl_lock(); + err = __tc_setup_cb_egdev_register(dev, cb, cb_priv); + rtnl_unlock(); + return err; +} +EXPORT_SYMBOL_GPL(tc_setup_cb_egdev_register); + +static void __tc_setup_cb_egdev_unregister(const struct net_device *dev, + tc_setup_cb_t *cb, void *cb_priv) +{ + struct tcf_action_egdev *egdev = tcf_action_egdev_lookup(dev); + + if (WARN_ON(!egdev)) + return; + tcf_action_egdev_cb_del(egdev, cb, cb_priv); + tcf_action_egdev_put(egdev); +} +void tc_setup_cb_egdev_unregister(const struct net_device *dev, + tc_setup_cb_t *cb, void *cb_priv) +{ + rtnl_lock(); + __tc_setup_cb_egdev_unregister(dev, cb, cb_priv); + rtnl_unlock(); +} +EXPORT_SYMBOL_GPL(tc_setup_cb_egdev_unregister); + +int tc_setup_cb_egdev_call(const struct net_device *dev, + enum tc_setup_type type, void *type_data, + bool err_stop) +{ + struct tcf_action_egdev *egdev = tcf_action_egdev_lookup(dev); + + if (!egdev) + return 0; + return tcf_action_egdev_cb_call(egdev, type, type_data, err_stop); +} +EXPORT_SYMBOL_GPL(tc_setup_cb_egdev_call); + +static __net_init int tcf_action_net_init(struct net *net) +{ + struct tcf_action_net *tan = net_generic(net, tcf_action_net_id); + + return rhashtable_init(&tan->egdev_ht, &tcf_action_egdev_ht_params); +} + +static void __net_exit tcf_action_net_exit(struct net *net) +{ + struct tcf_action_net *tan = net_generic(net, tcf_action_net_id); + + rhashtable_destroy(&tan->egdev_ht); +} + +static struct pernet_operations tcf_action_net_ops = { + .init = tcf_action_net_init, + .exit = tcf_action_net_exit, + .id = &tcf_action_net_id, + .size = sizeof(struct tcf_action_net), +}; + static int __init tc_action_init(void) { + int err; + + err = register_pernet_subsys(&tcf_action_net_ops); + if (err) + return err; + rtnl_register(PF_UNSPEC, RTM_NEWACTION, tc_ctl_action, NULL, 0); rtnl_register(PF_UNSPEC, RTM_DELACTION, tc_ctl_action, NULL, 0); rtnl_register(PF_UNSPEC, RTM_GETACTION, tc_ctl_action, tc_dump_action, diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 450873b0c4b9..99f9432f63cf 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -1026,6 +1026,36 @@ int tcf_exts_get_dev(struct net_device *dev, struct tcf_exts *exts, } EXPORT_SYMBOL(tcf_exts_get_dev); +int tcf_exts_egdev_cb_call(struct tcf_exts *exts, enum tc_setup_type type, + void *type_data, bool err_stop) +{ + int ok_count = 0; +#ifdef CONFIG_NET_CLS_ACT + const struct tc_action *a; + struct net_device *dev; + LIST_HEAD(actions); + int ret; + + if (!tcf_exts_has_actions(exts)) + return 0; + + tcf_exts_to_list(exts, &actions); + list_for_each_entry(a, &actions, list) { + if (!a->ops->get_dev) + continue; + dev = a->ops->get_dev(a); + if (!dev || !tc_can_offload(dev)) + continue; + ret = tc_setup_cb_egdev_call(dev, type, type_data, err_stop); + if (ret < 0) + return ret; + ok_count += ret; + } +#endif + return ok_count; +} +EXPORT_SYMBOL(tcf_exts_egdev_cb_call); + static int __init tc_filter_init(void) { rtnl_register(PF_UNSPEC, RTM_NEWTFILTER, tc_ctl_tfilter, NULL, 0); -- cgit v1.2.3 From 717503b9cf57c0bb7ea4d3a9f5699c9a04adf988 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Wed, 11 Oct 2017 09:41:09 +0200 Subject: net: sched: convert cls_flower->egress_dev users to tc_setup_cb_egdev infra The only user of cls_flower->egress_dev is mlx5. So do the conversion there alongside with the code originating the call in cls_flower function fl_hw_replace_filter to the newly introduced egress device callback infrastucture. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/en.h | 3 ++ drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 4 +- drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 31 +++++++---- include/net/pkt_cls.h | 5 +- net/sched/cls_api.c | 13 +++-- net/sched/cls_flower.c | 63 ++++++++++++----------- 6 files changed, 73 insertions(+), 46 deletions(-) (limited to 'include/net') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index cc13d3dbd366..5ec6d3e8dc89 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -1081,6 +1081,9 @@ int mlx5e_ethtool_get_ts_info(struct mlx5e_priv *priv, int mlx5e_ethtool_flash_device(struct mlx5e_priv *priv, struct ethtool_flash *flash); +int mlx5e_setup_tc(struct net_device *dev, enum tc_setup_type type, + void *type_data); + /* mlx5e generic netdev management API */ struct net_device* mlx5e_create_netdev(struct mlx5_core_dev *mdev, const struct mlx5e_profile *profile, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index cc11bbbd0309..2a32102e7648 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -3108,8 +3108,8 @@ static int mlx5e_setup_tc_cls_flower(struct net_device *dev, } #endif -static int mlx5e_setup_tc(struct net_device *dev, enum tc_setup_type type, - void *type_data) +int mlx5e_setup_tc(struct net_device *dev, enum tc_setup_type type, + void *type_data) { switch (type) { #ifdef CONFIG_MLX5_ESWITCH diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c index 45e03c427faf..765fc74fbb1b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -34,6 +34,7 @@ #include #include #include +#include #include #include @@ -667,14 +668,6 @@ mlx5e_rep_setup_tc_cls_flower(struct net_device *dev, cls_flower->common.chain_index) return -EOPNOTSUPP; - if (cls_flower->egress_dev) { - struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; - - dev = mlx5_eswitch_get_uplink_netdev(esw); - return dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSFLOWER, - cls_flower); - } - switch (cls_flower->command) { case TC_CLSFLOWER_REPLACE: return mlx5e_configure_flower(priv, cls_flower); @@ -698,6 +691,14 @@ static int mlx5e_rep_setup_tc(struct net_device *dev, enum tc_setup_type type, } } +static int mlx5e_rep_setup_tc_cb(enum tc_setup_type type, void *type_data, + void *cb_priv) +{ + struct net_device *dev = cb_priv; + + return mlx5e_setup_tc(dev, type, type_data); +} + bool mlx5e_is_uplink_rep(struct mlx5e_priv *priv) { struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; @@ -1017,15 +1018,24 @@ mlx5e_vport_rep_load(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep) goto err_detach_netdev; } + err = tc_setup_cb_egdev_register(netdev, mlx5e_rep_setup_tc_cb, + mlx5_eswitch_get_uplink_netdev(esw)); + if (err) + goto err_neigh_cleanup; + err = register_netdev(netdev); if (err) { pr_warn("Failed to register representor netdev for vport %d\n", rep->vport); - goto err_neigh_cleanup; + goto err_egdev_cleanup; } return 0; +err_egdev_cleanup: + tc_setup_cb_egdev_unregister(netdev, mlx5e_rep_setup_tc_cb, + mlx5_eswitch_get_uplink_netdev(esw)); + err_neigh_cleanup: mlx5e_rep_neigh_cleanup(rpriv); @@ -1047,7 +1057,8 @@ mlx5e_vport_rep_unload(struct mlx5_eswitch *esw, struct mlx5_eswitch_rep *rep) void *ppriv = priv->ppriv; unregister_netdev(rep->netdev); - + tc_setup_cb_egdev_unregister(netdev, mlx5e_rep_setup_tc_cb, + mlx5_eswitch_get_uplink_netdev(esw)); mlx5e_rep_neigh_cleanup(rpriv); mlx5e_detach_netdev(priv); mlx5e_destroy_netdev(priv); diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index 6f8149c82571..c0bdf5cad727 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -206,8 +206,6 @@ int tcf_exts_dump(struct sk_buff *skb, struct tcf_exts *exts); int tcf_exts_dump_stats(struct sk_buff *skb, struct tcf_exts *exts); int tcf_exts_get_dev(struct net_device *dev, struct tcf_exts *exts, struct net_device **hw_dev); -int tcf_exts_egdev_cb_call(struct tcf_exts *exts, enum tc_setup_type type, - void *type_data, bool err_stop); /** * struct tcf_pkt_info - packet information @@ -407,6 +405,9 @@ tcf_match_indev(struct sk_buff *skb, int ifindex) } #endif /* CONFIG_NET_CLS_IND */ +int tc_setup_cb_call(struct tcf_exts *exts, enum tc_setup_type type, + void *type_data, bool err_stop); + struct tc_cls_common_offload { u32 chain_index; __be16 protocol; diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 99f9432f63cf..51994a202585 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -1026,8 +1026,9 @@ int tcf_exts_get_dev(struct net_device *dev, struct tcf_exts *exts, } EXPORT_SYMBOL(tcf_exts_get_dev); -int tcf_exts_egdev_cb_call(struct tcf_exts *exts, enum tc_setup_type type, - void *type_data, bool err_stop) +static int tc_exts_setup_cb_egdev_call(struct tcf_exts *exts, + enum tc_setup_type type, + void *type_data, bool err_stop) { int ok_count = 0; #ifdef CONFIG_NET_CLS_ACT @@ -1054,7 +1055,13 @@ int tcf_exts_egdev_cb_call(struct tcf_exts *exts, enum tc_setup_type type, #endif return ok_count; } -EXPORT_SYMBOL(tcf_exts_egdev_cb_call); + +int tc_setup_cb_call(struct tcf_exts *exts, enum tc_setup_type type, + void *type_data, bool err_stop) +{ + return tc_exts_setup_cb_egdev_call(exts, type, type_data, err_stop); +} +EXPORT_SYMBOL(tc_setup_cb_call); static int __init tc_filter_init(void) { diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index db831ac708f6..5b7bb968d1d4 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -88,7 +88,6 @@ struct cls_fl_filter { u32 handle; u32 flags; struct rcu_head rcu; - struct net_device *hw_dev; }; static unsigned short int fl_mask_range(const struct fl_flow_mask *mask) @@ -201,16 +200,17 @@ static void fl_destroy_filter(struct rcu_head *head) static void fl_hw_destroy_filter(struct tcf_proto *tp, struct cls_fl_filter *f) { struct tc_cls_flower_offload cls_flower = {}; - struct net_device *dev = f->hw_dev; - - if (!tc_can_offload(dev)) - return; + struct net_device *dev = tp->q->dev_queue->dev; tc_cls_common_offload_init(&cls_flower.common, tp); cls_flower.command = TC_CLSFLOWER_DESTROY; cls_flower.cookie = (unsigned long) f; - dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSFLOWER, &cls_flower); + if (tc_can_offload(dev)) + dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSFLOWER, + &cls_flower); + tc_setup_cb_call(&f->exts, TC_SETUP_CLSFLOWER, + &cls_flower, false); } static int fl_hw_replace_filter(struct tcf_proto *tp, @@ -220,20 +220,9 @@ static int fl_hw_replace_filter(struct tcf_proto *tp, { struct net_device *dev = tp->q->dev_queue->dev; struct tc_cls_flower_offload cls_flower = {}; + bool skip_sw = tc_skip_sw(f->flags); int err; - if (!tc_can_offload(dev)) { - if (tcf_exts_get_dev(dev, &f->exts, &f->hw_dev) || - (f->hw_dev && !tc_can_offload(f->hw_dev))) { - f->hw_dev = dev; - return tc_skip_sw(f->flags) ? -EINVAL : 0; - } - dev = f->hw_dev; - cls_flower.egress_dev = true; - } else { - f->hw_dev = dev; - } - tc_cls_common_offload_init(&cls_flower.common, tp); cls_flower.command = TC_CLSFLOWER_REPLACE; cls_flower.cookie = (unsigned long) f; @@ -242,31 +231,47 @@ static int fl_hw_replace_filter(struct tcf_proto *tp, cls_flower.key = &f->mkey; cls_flower.exts = &f->exts; - err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSFLOWER, - &cls_flower); - if (!err) - f->flags |= TCA_CLS_FLAGS_IN_HW; + if (tc_can_offload(dev)) { + err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSFLOWER, + &cls_flower); + if (err) { + if (skip_sw) + return err; + } else { + f->flags |= TCA_CLS_FLAGS_IN_HW; + } + } - if (tc_skip_sw(f->flags)) + err = tc_setup_cb_call(&f->exts, TC_SETUP_CLSFLOWER, + &cls_flower, skip_sw); + if (err < 0) { + fl_hw_destroy_filter(tp, f); return err; + } else if (err > 0) { + f->flags |= TCA_CLS_FLAGS_IN_HW; + } + + if (skip_sw && !(f->flags & TCA_CLS_FLAGS_IN_HW)) + return -EINVAL; + return 0; } static void fl_hw_update_stats(struct tcf_proto *tp, struct cls_fl_filter *f) { struct tc_cls_flower_offload cls_flower = {}; - struct net_device *dev = f->hw_dev; - - if (!tc_can_offload(dev)) - return; + struct net_device *dev = tp->q->dev_queue->dev; tc_cls_common_offload_init(&cls_flower.common, tp); cls_flower.command = TC_CLSFLOWER_STATS; cls_flower.cookie = (unsigned long) f; cls_flower.exts = &f->exts; - dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSFLOWER, - &cls_flower); + if (tc_can_offload(dev)) + dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSFLOWER, + &cls_flower); + tc_setup_cb_call(&f->exts, TC_SETUP_CLSFLOWER, + &cls_flower, false); } static void __fl_delete(struct tcf_proto *tp, struct cls_fl_filter *f) -- cgit v1.2.3 From 7578d7b45ed870b13a8ace57e32feaed623c2a94 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Wed, 11 Oct 2017 09:41:10 +0200 Subject: net: sched: remove unused tcf_exts_get_dev helper and cls_flower->egress_dev The helper and the struct field ares no longer used by any code, so remove them. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/pkt_cls.h | 3 --- net/sched/cls_api.c | 22 ---------------------- 2 files changed, 25 deletions(-) (limited to 'include/net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index c0bdf5cad727..f5263743076b 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -204,8 +204,6 @@ void tcf_exts_destroy(struct tcf_exts *exts); void tcf_exts_change(struct tcf_exts *dst, struct tcf_exts *src); int tcf_exts_dump(struct sk_buff *skb, struct tcf_exts *exts); int tcf_exts_dump_stats(struct sk_buff *skb, struct tcf_exts *exts); -int tcf_exts_get_dev(struct net_device *dev, struct tcf_exts *exts, - struct net_device **hw_dev); /** * struct tcf_pkt_info - packet information @@ -517,7 +515,6 @@ struct tc_cls_flower_offload { struct fl_flow_key *mask; struct fl_flow_key *key; struct tcf_exts *exts; - bool egress_dev; }; enum tc_matchall_command { diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 51994a202585..2977b8a90851 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -1004,28 +1004,6 @@ int tcf_exts_dump_stats(struct sk_buff *skb, struct tcf_exts *exts) } EXPORT_SYMBOL(tcf_exts_dump_stats); -int tcf_exts_get_dev(struct net_device *dev, struct tcf_exts *exts, - struct net_device **hw_dev) -{ -#ifdef CONFIG_NET_CLS_ACT - const struct tc_action *a; - LIST_HEAD(actions); - - if (!tcf_exts_has_actions(exts)) - return -EINVAL; - - tcf_exts_to_list(exts, &actions); - list_for_each_entry(a, &actions, list) { - if (a->ops->get_dev) - *hw_dev = a->ops->get_dev(a); - } - if (*hw_dev) - return 0; -#endif - return -EOPNOTSUPP; -} -EXPORT_SYMBOL(tcf_exts_get_dev); - static int tc_exts_setup_cb_egdev_call(struct tcf_exts *exts, enum tc_setup_type type, void *type_data, bool err_stop) -- cgit v1.2.3 From 437d2762ba07f0fc639d5a09acb323fe4106a61f Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 11 Oct 2017 20:45:40 -0700 Subject: tcp: remove obsolete helpers Remove three inline helpers that are no longer needed. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/tcp.h | 17 ----------------- 1 file changed, 17 deletions(-) (limited to 'include/net') diff --git a/include/net/tcp.h b/include/net/tcp.h index 15163454174b..3b3b9b968e2d 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1629,18 +1629,6 @@ static inline struct sk_buff *tcp_write_queue_tail(const struct sock *sk) return skb_peek_tail(&sk->sk_write_queue); } -static inline struct sk_buff *tcp_write_queue_next(const struct sock *sk, - const struct sk_buff *skb) -{ - return skb_queue_next(&sk->sk_write_queue, skb); -} - -static inline struct sk_buff *tcp_write_queue_prev(const struct sock *sk, - const struct sk_buff *skb) -{ - return skb_queue_prev(&sk->sk_write_queue, skb); -} - #define tcp_for_write_queue_from_safe(skb, tmp, sk) \ skb_queue_walk_from_safe(&(sk)->sk_write_queue, skb, tmp) @@ -1697,11 +1685,6 @@ static inline void tcp_add_write_queue_tail(struct sock *sk, struct sk_buff *skb } } -static inline void __tcp_add_write_queue_head(struct sock *sk, struct sk_buff *skb) -{ - __skb_queue_head(&sk->sk_write_queue, skb); -} - /* Insert new before skb on the write queue of sk. */ static inline void tcp_insert_write_queue_before(struct sk_buff *new, struct sk_buff *skb, -- cgit v1.2.3 From 60724d4bae14cd295b27b1610cad9a2720eb0860 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Wed, 11 Oct 2017 10:57:48 -0700 Subject: net: dsa: Add support for DSA specific notifiers In preparation for communicating a given DSA network device's port number and switch index, create a specialized DSA notifier and two events: DSA_PORT_REGISTER and DSA_PORT_UNREGISTER that communicate: the slave network device (slave_dev), port number and switch number in the tree. This will be later used for network device drivers like bcmsysport which needs to cooperate with its DSA network devices to set-up queue mapping and scheduling. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- include/net/dsa.h | 45 +++++++++++++++++++++++++++++++++++++++++++++ net/dsa/dsa.c | 23 +++++++++++++++++++++++ net/dsa/slave.c | 13 +++++++++++++ 3 files changed, 81 insertions(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 10dceccd9ce8..40a709a0754d 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -471,4 +471,49 @@ static inline int dsa_switch_resume(struct dsa_switch *ds) } #endif /* CONFIG_PM_SLEEP */ +enum dsa_notifier_type { + DSA_PORT_REGISTER, + DSA_PORT_UNREGISTER, +}; + +struct dsa_notifier_info { + struct net_device *dev; +}; + +struct dsa_notifier_register_info { + struct dsa_notifier_info info; /* must be first */ + struct net_device *master; + unsigned int port_number; + unsigned int switch_number; +}; + +static inline struct net_device * +dsa_notifier_info_to_dev(const struct dsa_notifier_info *info) +{ + return info->dev; +} + +#if IS_ENABLED(CONFIG_NET_DSA) +int register_dsa_notifier(struct notifier_block *nb); +int unregister_dsa_notifier(struct notifier_block *nb); +int call_dsa_notifiers(unsigned long val, struct net_device *dev, + struct dsa_notifier_info *info); +#else +static inline int register_dsa_notifier(struct notifier_block *nb) +{ + return 0; +} + +static inline int unregister_dsa_notifier(struct notifier_block *nb) +{ + return 0; +} + +static inline int call_dsa_notifiers(unsigned long val, struct net_device *dev, + struct dsa_notifier_info *info) +{ + return NOTIFY_DONE; +} +#endif + #endif diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 51ca2a524a27..832c659ff993 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -261,6 +262,28 @@ bool dsa_schedule_work(struct work_struct *work) return queue_work(dsa_owq, work); } +static ATOMIC_NOTIFIER_HEAD(dsa_notif_chain); + +int register_dsa_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_register(&dsa_notif_chain, nb); +} +EXPORT_SYMBOL_GPL(register_dsa_notifier); + +int unregister_dsa_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_unregister(&dsa_notif_chain, nb); +} +EXPORT_SYMBOL_GPL(unregister_dsa_notifier); + +int call_dsa_notifiers(unsigned long val, struct net_device *dev, + struct dsa_notifier_info *info) +{ + info->dev = dev; + return atomic_notifier_call_chain(&dsa_notif_chain, val, info); +} +EXPORT_SYMBOL_GPL(call_dsa_notifiers); + static int __init dsa_init_module(void) { int rc; diff --git a/net/dsa/slave.c b/net/dsa/slave.c index fb2954ff198c..45f4ea845c07 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -1116,6 +1116,7 @@ int dsa_slave_resume(struct net_device *slave_dev) int dsa_slave_create(struct dsa_port *port, const char *name) { + struct dsa_notifier_register_info rinfo = { }; struct dsa_switch *ds = port->ds; struct net_device *master; struct net_device *slave_dev; @@ -1177,6 +1178,12 @@ int dsa_slave_create(struct dsa_port *port, const char *name) goto out_free; } + rinfo.info.dev = slave_dev; + rinfo.master = master; + rinfo.port_number = p->dp->index; + rinfo.switch_number = p->dp->ds->index; + call_dsa_notifiers(DSA_PORT_REGISTER, slave_dev, &rinfo.info); + ret = register_netdev(slave_dev); if (ret) { netdev_err(master, "error %d registering interface %s\n", @@ -1200,6 +1207,7 @@ out_free: void dsa_slave_destroy(struct net_device *slave_dev) { struct dsa_slave_priv *p = netdev_priv(slave_dev); + struct dsa_notifier_register_info rinfo = { }; struct device_node *port_dn; port_dn = p->dp->dn; @@ -1211,6 +1219,11 @@ void dsa_slave_destroy(struct net_device *slave_dev) if (of_phy_is_fixed_link(port_dn)) of_phy_deregister_fixed_link(port_dn); } + rinfo.info.dev = slave_dev; + rinfo.master = p->dp->cpu_dp->netdev; + rinfo.port_number = p->dp->index; + rinfo.switch_number = p->dp->ds->index; + call_dsa_notifiers(DSA_PORT_UNREGISTER, slave_dev, &rinfo.info); unregister_netdev(slave_dev); free_percpu(p->stats64); free_netdev(slave_dev); -- cgit v1.2.3 From 0a5f14ce67a6e093e651d3cd75e6ac281123d93a Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Wed, 11 Oct 2017 10:57:49 -0700 Subject: net: dsa: tag_brcm: Indicate to master netdevice port + queue We need to tell the DSA master network device doing the actual transmission what the desired switch port and queue number is for it to resolve that to the internal transmit queue it is mapped to. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- include/net/dsa.h | 5 +++++ net/dsa/tag_brcm.c | 6 ++++++ 2 files changed, 11 insertions(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 40a709a0754d..ce1d622734d7 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -516,4 +516,9 @@ static inline int call_dsa_notifiers(unsigned long val, struct net_device *dev, } #endif +/* Broadcom tag specific helpers to insert and extract queue/port number */ +#define BRCM_TAG_SET_PORT_QUEUE(p, q) ((p) << 8 | q) +#define BRCM_TAG_GET_PORT(v) ((v) >> 8) +#define BRCM_TAG_GET_QUEUE(v) ((v) & 0xff) + #endif diff --git a/net/dsa/tag_brcm.c b/net/dsa/tag_brcm.c index 8e4bdb9d9ae3..cc4f472fbd77 100644 --- a/net/dsa/tag_brcm.c +++ b/net/dsa/tag_brcm.c @@ -86,6 +86,12 @@ static struct sk_buff *brcm_tag_xmit(struct sk_buff *skb, struct net_device *dev brcm_tag[2] = BRCM_IG_DSTMAP2_MASK; brcm_tag[3] = (1 << p->dp->index) & BRCM_IG_DSTMAP1_MASK; + /* Now tell the master network device about the desired output queue + * as well + */ + skb_set_queue_mapping(skb, BRCM_TAG_SET_PORT_QUEUE(p->dp->index, + queue)); + return skb; } -- cgit v1.2.3 From 8f04748016f3b583e675e0f649d42cfc10812a8b Mon Sep 17 00:00:00 2001 From: Roman Mashak Date: Wed, 11 Oct 2017 10:50:29 -0400 Subject: net sched actions: change IFE modules alias names Make style of module alias name consistent with other subsystems in kernel, for example net devices. Fixes: 084e2f6566d2 ("Support to encoding decoding skb mark on IFE action") Fixes: 200e10f46936 ("Support to encoding decoding skb prio on IFE action") Fixes: 408fbc22ef1e ("net sched ife action: Introduce skb tcindex metadata encap decap") Signed-off-by: Roman Mashak Signed-off-by: David S. Miller --- include/net/tc_act/tc_ife.h | 2 +- net/sched/act_ife.c | 2 +- net/sched/act_meta_mark.c | 2 +- net/sched/act_meta_skbprio.c | 2 +- net/sched/act_meta_skbtcindex.c | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) (limited to 'include/net') diff --git a/include/net/tc_act/tc_ife.h b/include/net/tc_act/tc_ife.h index 30ba459ddd34..104578f16062 100644 --- a/include/net/tc_act/tc_ife.h +++ b/include/net/tc_act/tc_ife.h @@ -40,7 +40,7 @@ struct tcf_meta_ops { struct module *owner; }; -#define MODULE_ALIAS_IFE_META(metan) MODULE_ALIAS("ifemeta" __stringify_1(metan)) +#define MODULE_ALIAS_IFE_META(metan) MODULE_ALIAS("ife-meta-" metan) int ife_get_meta_u32(struct sk_buff *skb, struct tcf_meta_info *mi); int ife_get_meta_u16(struct sk_buff *skb, struct tcf_meta_info *mi); diff --git a/net/sched/act_ife.c b/net/sched/act_ife.c index 8ccd35825b6b..791aeee11c7e 100644 --- a/net/sched/act_ife.c +++ b/net/sched/act_ife.c @@ -263,7 +263,7 @@ static int load_metaops_and_vet(struct tcf_ife_info *ife, u32 metaid, if (exists) spin_unlock_bh(&ife->tcf_lock); rtnl_unlock(); - request_module("ifemeta%u", metaid); + request_module("ife-meta-%u", metaid); rtnl_lock(); if (exists) spin_lock_bh(&ife->tcf_lock); diff --git a/net/sched/act_meta_mark.c b/net/sched/act_meta_mark.c index 82892170ce4f..1e3f10e5da99 100644 --- a/net/sched/act_meta_mark.c +++ b/net/sched/act_meta_mark.c @@ -76,4 +76,4 @@ module_exit(ifemark_cleanup_module); MODULE_AUTHOR("Jamal Hadi Salim(2015)"); MODULE_DESCRIPTION("Inter-FE skb mark metadata module"); MODULE_LICENSE("GPL"); -MODULE_ALIAS_IFE_META(IFE_META_SKBMARK); +MODULE_ALIAS_IFE_META("skbmark"); diff --git a/net/sched/act_meta_skbprio.c b/net/sched/act_meta_skbprio.c index 26bf4d86030b..4033f9fc4d4a 100644 --- a/net/sched/act_meta_skbprio.c +++ b/net/sched/act_meta_skbprio.c @@ -73,4 +73,4 @@ module_exit(ifeprio_cleanup_module); MODULE_AUTHOR("Jamal Hadi Salim(2015)"); MODULE_DESCRIPTION("Inter-FE skb prio metadata action"); MODULE_LICENSE("GPL"); -MODULE_ALIAS_IFE_META(IFE_META_PRIO); +MODULE_ALIAS_IFE_META("skbprio"); diff --git a/net/sched/act_meta_skbtcindex.c b/net/sched/act_meta_skbtcindex.c index 3b35774ce890..2ea1f26c9e96 100644 --- a/net/sched/act_meta_skbtcindex.c +++ b/net/sched/act_meta_skbtcindex.c @@ -76,4 +76,4 @@ module_exit(ifetc_index_cleanup_module); MODULE_AUTHOR("Jamal Hadi Salim(2016)"); MODULE_DESCRIPTION("Inter-FE skb tc_index metadata module"); MODULE_LICENSE("GPL"); -MODULE_ALIAS_IFE_META(IFE_META_SKBTCINDEX); +MODULE_ALIAS_IFE_META("tcindex"); -- cgit v1.2.3 From aa9fd9a325d51fa0b11153b03b8fefff569fa955 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 11 Oct 2017 17:16:08 -0400 Subject: sched: act: ife: update parameters via rcu handling This patch changes the parameter updating via RCU and not protected by a spinlock anymore. This reduce the time that the spinlock is being held. Signed-off-by: Alexander Aring Acked-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- include/net/tc_act/tc_ife.h | 10 ++++-- net/sched/act_ife.c | 87 ++++++++++++++++++++++++++++++--------------- 2 files changed, 67 insertions(+), 30 deletions(-) (limited to 'include/net') diff --git a/include/net/tc_act/tc_ife.h b/include/net/tc_act/tc_ife.h index 104578f16062..c7fb99c3f76c 100644 --- a/include/net/tc_act/tc_ife.h +++ b/include/net/tc_act/tc_ife.h @@ -6,12 +6,18 @@ #include #include -struct tcf_ife_info { - struct tc_action common; +struct tcf_ife_params { u8 eth_dst[ETH_ALEN]; u8 eth_src[ETH_ALEN]; u16 eth_type; u16 flags; + + struct rcu_head rcu; +}; + +struct tcf_ife_info { + struct tc_action common; + struct tcf_ife_params __rcu *params; /* list of metaids allowed */ struct list_head metalist; }; diff --git a/net/sched/act_ife.c b/net/sched/act_ife.c index f59d78918cf9..252ee7d8c731 100644 --- a/net/sched/act_ife.c +++ b/net/sched/act_ife.c @@ -406,10 +406,14 @@ static void _tcf_ife_cleanup(struct tc_action *a, int bind) static void tcf_ife_cleanup(struct tc_action *a, int bind) { struct tcf_ife_info *ife = to_ife(a); + struct tcf_ife_params *p; spin_lock_bh(&ife->tcf_lock); _tcf_ife_cleanup(a, bind); spin_unlock_bh(&ife->tcf_lock); + + p = rcu_dereference_protected(ife->params, 1); + kfree_rcu(p, rcu); } /* under ife->tcf_lock for existing action */ @@ -446,6 +450,7 @@ static int tcf_ife_init(struct net *net, struct nlattr *nla, struct tc_action_net *tn = net_generic(net, ife_net_id); struct nlattr *tb[TCA_IFE_MAX + 1]; struct nlattr *tb2[IFE_META_MAX + 1]; + struct tcf_ife_params *p, *p_old; struct tcf_ife_info *ife; u16 ife_type = ETH_P_IFE; struct tc_ife *parm; @@ -471,24 +476,34 @@ static int tcf_ife_init(struct net *net, struct nlattr *nla, if (parm->flags & ~IFE_ENCODE) return -EINVAL; + p = kzalloc(sizeof(*p), GFP_KERNEL); + if (!p) + return -ENOMEM; + exists = tcf_idr_check(tn, parm->index, a, bind); - if (exists && bind) + if (exists && bind) { + kfree(p); return 0; + } if (!exists) { ret = tcf_idr_create(tn, parm->index, est, a, &act_ife_ops, bind, true); - if (ret) + if (ret) { + kfree(p); return ret; + } ret = ACT_P_CREATED; } else { tcf_idr_release(*a, bind); - if (!ovr) + if (!ovr) { + kfree(p); return -EEXIST; + } } ife = to_ife(*a); - ife->flags = parm->flags; + p->flags = parm->flags; if (parm->flags & IFE_ENCODE) { if (tb[TCA_IFE_TYPE]) @@ -499,24 +514,25 @@ static int tcf_ife_init(struct net *net, struct nlattr *nla, saddr = nla_data(tb[TCA_IFE_SMAC]); } - if (exists) - spin_lock_bh(&ife->tcf_lock); ife->tcf_action = parm->action; if (parm->flags & IFE_ENCODE) { if (daddr) - ether_addr_copy(ife->eth_dst, daddr); + ether_addr_copy(p->eth_dst, daddr); else - eth_zero_addr(ife->eth_dst); + eth_zero_addr(p->eth_dst); if (saddr) - ether_addr_copy(ife->eth_src, saddr); + ether_addr_copy(p->eth_src, saddr); else - eth_zero_addr(ife->eth_src); + eth_zero_addr(p->eth_src); - ife->eth_type = ife_type; + p->eth_type = ife_type; } + if (exists) + spin_lock_bh(&ife->tcf_lock); + if (ret == ACT_P_CREATED) INIT_LIST_HEAD(&ife->metalist); @@ -532,6 +548,7 @@ metadata_parse_err: if (exists) spin_unlock_bh(&ife->tcf_lock); + kfree(p); return err; } @@ -552,6 +569,7 @@ metadata_parse_err: if (exists) spin_unlock_bh(&ife->tcf_lock); + kfree(p); return err; } } @@ -559,6 +577,11 @@ metadata_parse_err: if (exists) spin_unlock_bh(&ife->tcf_lock); + p_old = rtnl_dereference(ife->params); + rcu_assign_pointer(ife->params, p); + if (p_old) + kfree_rcu(p_old, rcu); + if (ret == ACT_P_CREATED) tcf_idr_insert(tn, *a); @@ -570,12 +593,13 @@ static int tcf_ife_dump(struct sk_buff *skb, struct tc_action *a, int bind, { unsigned char *b = skb_tail_pointer(skb); struct tcf_ife_info *ife = to_ife(a); + struct tcf_ife_params *p = rtnl_dereference(ife->params); struct tc_ife opt = { .index = ife->tcf_index, .refcnt = ife->tcf_refcnt - ref, .bindcnt = ife->tcf_bindcnt - bind, .action = ife->tcf_action, - .flags = ife->flags, + .flags = p->flags, }; struct tcf_t t; @@ -586,17 +610,17 @@ static int tcf_ife_dump(struct sk_buff *skb, struct tc_action *a, int bind, if (nla_put_64bit(skb, TCA_IFE_TM, sizeof(t), &t, TCA_IFE_PAD)) goto nla_put_failure; - if (!is_zero_ether_addr(ife->eth_dst)) { - if (nla_put(skb, TCA_IFE_DMAC, ETH_ALEN, ife->eth_dst)) + if (!is_zero_ether_addr(p->eth_dst)) { + if (nla_put(skb, TCA_IFE_DMAC, ETH_ALEN, p->eth_dst)) goto nla_put_failure; } - if (!is_zero_ether_addr(ife->eth_src)) { - if (nla_put(skb, TCA_IFE_SMAC, ETH_ALEN, ife->eth_src)) + if (!is_zero_ether_addr(p->eth_src)) { + if (nla_put(skb, TCA_IFE_SMAC, ETH_ALEN, p->eth_src)) goto nla_put_failure; } - if (nla_put(skb, TCA_IFE_TYPE, 2, &ife->eth_type)) + if (nla_put(skb, TCA_IFE_TYPE, 2, &p->eth_type)) goto nla_put_failure; if (dump_metalist(skb, ife)) { @@ -698,7 +722,7 @@ static int ife_get_sz(struct sk_buff *skb, struct tcf_ife_info *ife) } static int tcf_ife_encode(struct sk_buff *skb, const struct tc_action *a, - struct tcf_result *res) + struct tcf_result *res, struct tcf_ife_params *p) { struct tcf_ife_info *ife = to_ife(a); int action = ife->tcf_action; @@ -762,19 +786,18 @@ static int tcf_ife_encode(struct sk_buff *skb, const struct tc_action *a, } skboff += err; } + spin_unlock(&ife->tcf_lock); oethh = (struct ethhdr *)skb->data; - if (!is_zero_ether_addr(ife->eth_src)) - ether_addr_copy(oethh->h_source, ife->eth_src); - if (!is_zero_ether_addr(ife->eth_dst)) - ether_addr_copy(oethh->h_dest, ife->eth_dst); - oethh->h_proto = htons(ife->eth_type); + if (!is_zero_ether_addr(p->eth_src)) + ether_addr_copy(oethh->h_source, p->eth_src); + if (!is_zero_ether_addr(p->eth_dst)) + ether_addr_copy(oethh->h_dest, p->eth_dst); + oethh->h_proto = htons(p->eth_type); if (skb_at_tc_ingress(skb)) skb_pull(skb, skb->dev->hard_header_len); - spin_unlock(&ife->tcf_lock); - return action; } @@ -782,9 +805,17 @@ static int tcf_ife_act(struct sk_buff *skb, const struct tc_action *a, struct tcf_result *res) { struct tcf_ife_info *ife = to_ife(a); - - if (ife->flags & IFE_ENCODE) - return tcf_ife_encode(skb, a, res); + struct tcf_ife_params *p; + int ret; + + rcu_read_lock(); + p = rcu_dereference(ife->params); + if (p->flags & IFE_ENCODE) { + ret = tcf_ife_encode(skb, a, res, p); + rcu_read_unlock(); + return ret; + } + rcu_read_unlock(); return tcf_ife_decode(skb, a, res); } -- cgit v1.2.3 From 4e8b86c062695454df0b76f3fee4fab8dc4bb716 Mon Sep 17 00:00:00 2001 From: Amritha Nambiar Date: Thu, 7 Sep 2017 04:00:06 -0700 Subject: mqprio: Introduce new hardware offload mode and shaper in mqprio The offload types currently supported in mqprio are 0 (no offload) and 1 (offload only TCs) by setting these values for the 'hw' option. If offloads are supported by setting the 'hw' option to 1, the default offload mode is 'dcb' where only the TC values are offloaded to the device. This patch introduces a new hardware offload mode called 'channel' with 'hw' set to 1 in mqprio which makes full use of the mqprio options, the TCs, the queue configurations and the QoS parameters for the TCs. This is achieved through a new netlink attribute for the 'mode' option which takes values such as 'dcb' (default) and 'channel'. The 'channel' mode also supports QoS attributes for traffic class such as minimum and maximum values for bandwidth rate limits. This patch enables configuring additional HW shaper attributes associated with a traffic class. Currently the shaper for bandwidth rate limiting is supported which takes options such as minimum and maximum bandwidth rates and are offloaded to the hardware in the 'channel' mode. The min and max limits for bandwidth rates are provided by the user along with the TCs and the queue configurations when creating the mqprio qdisc. The interface can be extended to support new HW shapers in future through the 'shaper' attribute. Introduces a new data structure 'tc_mqprio_qopt_offload' for offloading mqprio queue options and use this to be shared between the kernel and device driver. This contains a copy of the existing data structure for mqprio queue options. This new data structure can be extended when adding new attributes for traffic class such as mode, shaper, shaper parameters (bandwidth rate limits). The existing data structure for mqprio queue options will be shared between the kernel and userspace. Example: queues 4@0 4@4 hw 1 mode channel shaper bw_rlimit\ min_rate 1Gbit 2Gbit max_rate 4Gbit 5Gbit To dump the bandwidth rates: qdisc mqprio 804a: root tc 2 map 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 queues:(0:3) (4:7) mode:channel shaper:bw_rlimit min_rate:1Gbit 2Gbit max_rate:4Gbit 5Gbit Signed-off-by: Amritha Nambiar Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- include/net/pkt_cls.h | 9 ++ include/uapi/linux/pkt_sched.h | 32 +++++++ net/sched/sch_mqprio.c | 183 +++++++++++++++++++++++++++++++++++++++-- 3 files changed, 215 insertions(+), 9 deletions(-) (limited to 'include/net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index f5263743076b..60d39789e4f0 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -546,6 +546,15 @@ struct tc_cls_bpf_offload { u32 gen_flags; }; +struct tc_mqprio_qopt_offload { + /* struct tc_mqprio_qopt must always be the first element */ + struct tc_mqprio_qopt qopt; + u16 mode; + u16 shaper; + u32 flags; + u64 min_rate[TC_QOPT_MAX_QUEUE]; + u64 max_rate[TC_QOPT_MAX_QUEUE]; +}; /* This structure holds cookie structure that is passed from user * to the kernel for actions and classifiers diff --git a/include/uapi/linux/pkt_sched.h b/include/uapi/linux/pkt_sched.h index 099bf5528fed..e95b5c9b9fad 100644 --- a/include/uapi/linux/pkt_sched.h +++ b/include/uapi/linux/pkt_sched.h @@ -625,6 +625,22 @@ enum { #define TC_MQPRIO_HW_OFFLOAD_MAX (__TC_MQPRIO_HW_OFFLOAD_MAX - 1) +enum { + TC_MQPRIO_MODE_DCB, + TC_MQPRIO_MODE_CHANNEL, + __TC_MQPRIO_MODE_MAX +}; + +#define __TC_MQPRIO_MODE_MAX (__TC_MQPRIO_MODE_MAX - 1) + +enum { + TC_MQPRIO_SHAPER_DCB, + TC_MQPRIO_SHAPER_BW_RATE, /* Add new shapers below */ + __TC_MQPRIO_SHAPER_MAX +}; + +#define __TC_MQPRIO_SHAPER_MAX (__TC_MQPRIO_SHAPER_MAX - 1) + struct tc_mqprio_qopt { __u8 num_tc; __u8 prio_tc_map[TC_QOPT_BITMASK + 1]; @@ -633,6 +649,22 @@ struct tc_mqprio_qopt { __u16 offset[TC_QOPT_MAX_QUEUE]; }; +#define TC_MQPRIO_F_MODE 0x1 +#define TC_MQPRIO_F_SHAPER 0x2 +#define TC_MQPRIO_F_MIN_RATE 0x4 +#define TC_MQPRIO_F_MAX_RATE 0x8 + +enum { + TCA_MQPRIO_UNSPEC, + TCA_MQPRIO_MODE, + TCA_MQPRIO_SHAPER, + TCA_MQPRIO_MIN_RATE64, + TCA_MQPRIO_MAX_RATE64, + __TCA_MQPRIO_MAX, +}; + +#define TCA_MQPRIO_MAX (__TCA_MQPRIO_MAX - 1) + /* SFB */ enum { diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c index 6bcdfe6e7b63..f1ae9be83934 100644 --- a/net/sched/sch_mqprio.c +++ b/net/sched/sch_mqprio.c @@ -18,10 +18,16 @@ #include #include #include +#include struct mqprio_sched { struct Qdisc **qdiscs; + u16 mode; + u16 shaper; int hw_offload; + u32 flags; + u64 min_rate[TC_QOPT_MAX_QUEUE]; + u64 max_rate[TC_QOPT_MAX_QUEUE]; }; static void mqprio_destroy(struct Qdisc *sch) @@ -39,9 +45,17 @@ static void mqprio_destroy(struct Qdisc *sch) } if (priv->hw_offload && dev->netdev_ops->ndo_setup_tc) { - struct tc_mqprio_qopt mqprio = {}; + struct tc_mqprio_qopt_offload mqprio = { { 0 } }; - dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_MQPRIO, &mqprio); + switch (priv->mode) { + case TC_MQPRIO_MODE_DCB: + case TC_MQPRIO_MODE_CHANNEL: + dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_MQPRIO, + &mqprio); + break; + default: + return; + } } else { netdev_set_num_tc(dev, 0); } @@ -97,6 +111,26 @@ static int mqprio_parse_opt(struct net_device *dev, struct tc_mqprio_qopt *qopt) return 0; } +static const struct nla_policy mqprio_policy[TCA_MQPRIO_MAX + 1] = { + [TCA_MQPRIO_MODE] = { .len = sizeof(u16) }, + [TCA_MQPRIO_SHAPER] = { .len = sizeof(u16) }, + [TCA_MQPRIO_MIN_RATE64] = { .type = NLA_NESTED }, + [TCA_MQPRIO_MAX_RATE64] = { .type = NLA_NESTED }, +}; + +static int parse_attr(struct nlattr *tb[], int maxtype, struct nlattr *nla, + const struct nla_policy *policy, int len) +{ + int nested_len = nla_len(nla) - NLA_ALIGN(len); + + if (nested_len >= nla_attr_size(0)) + return nla_parse(tb, maxtype, nla_data(nla) + NLA_ALIGN(len), + nested_len, policy, NULL); + + memset(tb, 0, sizeof(struct nlattr *) * (maxtype + 1)); + return 0; +} + static int mqprio_init(struct Qdisc *sch, struct nlattr *opt) { struct net_device *dev = qdisc_dev(sch); @@ -105,6 +139,10 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt) struct Qdisc *qdisc; int i, err = -EOPNOTSUPP; struct tc_mqprio_qopt *qopt = NULL; + struct nlattr *tb[TCA_MQPRIO_MAX + 1]; + struct nlattr *attr; + int rem; + int len = nla_len(opt) - NLA_ALIGN(sizeof(*qopt)); BUILD_BUG_ON(TC_MAX_QUEUE != TC_QOPT_MAX_QUEUE); BUILD_BUG_ON(TC_BITMASK != TC_QOPT_BITMASK); @@ -122,6 +160,58 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt) if (mqprio_parse_opt(dev, qopt)) return -EINVAL; + if (len > 0) { + err = parse_attr(tb, TCA_MQPRIO_MAX, opt, mqprio_policy, + sizeof(*qopt)); + if (err < 0) + return err; + + if (!qopt->hw) + return -EINVAL; + + if (tb[TCA_MQPRIO_MODE]) { + priv->flags |= TC_MQPRIO_F_MODE; + priv->mode = *(u16 *)nla_data(tb[TCA_MQPRIO_MODE]); + } + + if (tb[TCA_MQPRIO_SHAPER]) { + priv->flags |= TC_MQPRIO_F_SHAPER; + priv->shaper = *(u16 *)nla_data(tb[TCA_MQPRIO_SHAPER]); + } + + if (tb[TCA_MQPRIO_MIN_RATE64]) { + if (priv->shaper != TC_MQPRIO_SHAPER_BW_RATE) + return -EINVAL; + i = 0; + nla_for_each_nested(attr, tb[TCA_MQPRIO_MIN_RATE64], + rem) { + if (nla_type(attr) != TCA_MQPRIO_MIN_RATE64) + return -EINVAL; + if (i >= qopt->num_tc) + break; + priv->min_rate[i] = *(u64 *)nla_data(attr); + i++; + } + priv->flags |= TC_MQPRIO_F_MIN_RATE; + } + + if (tb[TCA_MQPRIO_MAX_RATE64]) { + if (priv->shaper != TC_MQPRIO_SHAPER_BW_RATE) + return -EINVAL; + i = 0; + nla_for_each_nested(attr, tb[TCA_MQPRIO_MAX_RATE64], + rem) { + if (nla_type(attr) != TCA_MQPRIO_MAX_RATE64) + return -EINVAL; + if (i >= qopt->num_tc) + break; + priv->max_rate[i] = *(u64 *)nla_data(attr); + i++; + } + priv->flags |= TC_MQPRIO_F_MAX_RATE; + } + } + /* pre-allocate qdisc, attachment can't fail */ priv->qdiscs = kcalloc(dev->num_tx_queues, sizeof(priv->qdiscs[0]), GFP_KERNEL); @@ -146,14 +236,36 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt) * supplied and verified mapping */ if (qopt->hw) { - struct tc_mqprio_qopt mqprio = *qopt; + struct tc_mqprio_qopt_offload mqprio = {.qopt = *qopt}; - err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_MQPRIO, + switch (priv->mode) { + case TC_MQPRIO_MODE_DCB: + if (priv->shaper != TC_MQPRIO_SHAPER_DCB) + return -EINVAL; + break; + case TC_MQPRIO_MODE_CHANNEL: + mqprio.flags = priv->flags; + if (priv->flags & TC_MQPRIO_F_MODE) + mqprio.mode = priv->mode; + if (priv->flags & TC_MQPRIO_F_SHAPER) + mqprio.shaper = priv->shaper; + if (priv->flags & TC_MQPRIO_F_MIN_RATE) + for (i = 0; i < mqprio.qopt.num_tc; i++) + mqprio.min_rate[i] = priv->min_rate[i]; + if (priv->flags & TC_MQPRIO_F_MAX_RATE) + for (i = 0; i < mqprio.qopt.num_tc; i++) + mqprio.max_rate[i] = priv->max_rate[i]; + break; + default: + return -EINVAL; + } + err = dev->netdev_ops->ndo_setup_tc(dev, + TC_SETUP_MQPRIO, &mqprio); if (err) return err; - priv->hw_offload = mqprio.hw; + priv->hw_offload = mqprio.qopt.hw; } else { netdev_set_num_tc(dev, qopt->num_tc); for (i = 0; i < qopt->num_tc; i++) @@ -223,11 +335,51 @@ static int mqprio_graft(struct Qdisc *sch, unsigned long cl, struct Qdisc *new, return 0; } +static int dump_rates(struct mqprio_sched *priv, + struct tc_mqprio_qopt *opt, struct sk_buff *skb) +{ + struct nlattr *nest; + int i; + + if (priv->flags & TC_MQPRIO_F_MIN_RATE) { + nest = nla_nest_start(skb, TCA_MQPRIO_MIN_RATE64); + if (!nest) + goto nla_put_failure; + + for (i = 0; i < opt->num_tc; i++) { + if (nla_put(skb, TCA_MQPRIO_MIN_RATE64, + sizeof(priv->min_rate[i]), + &priv->min_rate[i])) + goto nla_put_failure; + } + nla_nest_end(skb, nest); + } + + if (priv->flags & TC_MQPRIO_F_MAX_RATE) { + nest = nla_nest_start(skb, TCA_MQPRIO_MAX_RATE64); + if (!nest) + goto nla_put_failure; + + for (i = 0; i < opt->num_tc; i++) { + if (nla_put(skb, TCA_MQPRIO_MAX_RATE64, + sizeof(priv->max_rate[i]), + &priv->max_rate[i])) + goto nla_put_failure; + } + nla_nest_end(skb, nest); + } + return 0; + +nla_put_failure: + nla_nest_cancel(skb, nest); + return -1; +} + static int mqprio_dump(struct Qdisc *sch, struct sk_buff *skb) { struct net_device *dev = qdisc_dev(sch); struct mqprio_sched *priv = qdisc_priv(sch); - unsigned char *b = skb_tail_pointer(skb); + struct nlattr *nla = (struct nlattr *)skb_tail_pointer(skb); struct tc_mqprio_qopt opt = { 0 }; struct Qdisc *qdisc; unsigned int i; @@ -258,12 +410,25 @@ static int mqprio_dump(struct Qdisc *sch, struct sk_buff *skb) opt.offset[i] = dev->tc_to_txq[i].offset; } - if (nla_put(skb, TCA_OPTIONS, sizeof(opt), &opt)) + if (nla_put(skb, TCA_OPTIONS, NLA_ALIGN(sizeof(opt)), &opt)) + goto nla_put_failure; + + if ((priv->flags & TC_MQPRIO_F_MODE) && + nla_put_u16(skb, TCA_MQPRIO_MODE, priv->mode)) + goto nla_put_failure; + + if ((priv->flags & TC_MQPRIO_F_SHAPER) && + nla_put_u16(skb, TCA_MQPRIO_SHAPER, priv->shaper)) + goto nla_put_failure; + + if ((priv->flags & TC_MQPRIO_F_MIN_RATE || + priv->flags & TC_MQPRIO_F_MAX_RATE) && + (dump_rates(priv, &opt, skb) != 0)) goto nla_put_failure; - return skb->len; + return nla_nest_end(skb, nla); nla_put_failure: - nlmsg_trim(skb, b); + nlmsg_trim(skb, nla); return -1; } -- cgit v1.2.3 From 841f4f24053acad69240c6ab7427a1d24bc29491 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Fri, 13 Oct 2017 14:18:09 -0400 Subject: net: dsa: remove .set_addr Now that there is no user for the .set_addr function, remove it from DSA. If a switch supports this feature (like mv88e6xxx), the implementation can be done in the driver setup. Signed-off-by: Vivien Didelot Signed-off-by: David S. Miller --- include/net/dsa.h | 1 - net/dsa/dsa2.c | 6 ------ net/dsa/legacy.c | 6 ------ 3 files changed, 13 deletions(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index ce1d622734d7..2746741f74cf 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -291,7 +291,6 @@ struct dsa_switch_ops { enum dsa_tag_protocol (*get_tag_protocol)(struct dsa_switch *ds); int (*setup)(struct dsa_switch *ds); - int (*set_addr)(struct dsa_switch *ds, u8 *addr); u32 (*get_phy_flags)(struct dsa_switch *ds, int port); /* diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index 54ed054777bd..6ac9e11d385c 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -336,12 +336,6 @@ static int dsa_ds_apply(struct dsa_switch_tree *dst, struct dsa_switch *ds) if (err) return err; - if (ds->ops->set_addr) { - err = ds->ops->set_addr(ds, dst->cpu_dp->netdev->dev_addr); - if (err < 0) - return err; - } - if (!ds->slave_mii_bus && ds->ops->phy_read) { ds->slave_mii_bus = devm_mdiobus_alloc(ds->dev); if (!ds->slave_mii_bus) diff --git a/net/dsa/legacy.c b/net/dsa/legacy.c index 19ff6e0a21dc..b0fefbffe082 100644 --- a/net/dsa/legacy.c +++ b/net/dsa/legacy.c @@ -172,12 +172,6 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, if (ret) return ret; - if (ops->set_addr) { - ret = ops->set_addr(ds, master->dev_addr); - if (ret < 0) - return ret; - } - if (!ds->slave_mii_bus && ops->phy_read) { ds->slave_mii_bus = devm_mdiobus_alloc(ds->dev); if (!ds->slave_mii_bus) -- cgit v1.2.3 From 69d78ef25c7b0058674145500efb12255738ba8a Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Fri, 13 Oct 2017 14:00:57 +0200 Subject: net: sched: store Qdisc pointer in struct block Prepare for removal of tp->q and store Qdisc pointer in the block structure. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/pkt_cls.h | 4 ++-- include/net/sch_generic.h | 1 + net/sched/cls_api.c | 3 ++- net/sched/sch_atm.c | 4 ++-- net/sched/sch_cbq.c | 2 +- net/sched/sch_drr.c | 2 +- net/sched/sch_dsmark.c | 2 +- net/sched/sch_fq_codel.c | 2 +- net/sched/sch_hfsc.c | 4 ++-- net/sched/sch_htb.c | 4 ++-- net/sched/sch_ingress.c | 6 +++--- net/sched/sch_multiq.c | 2 +- net/sched/sch_prio.c | 2 +- net/sched/sch_qfq.c | 2 +- net/sched/sch_sfb.c | 2 +- net/sched/sch_sfq.c | 2 +- 16 files changed, 23 insertions(+), 21 deletions(-) (limited to 'include/net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index 60d39789e4f0..e6c9e1e4d711 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -22,7 +22,7 @@ struct tcf_chain *tcf_chain_get(struct tcf_block *block, u32 chain_index, bool create); void tcf_chain_put(struct tcf_chain *chain); int tcf_block_get(struct tcf_block **p_block, - struct tcf_proto __rcu **p_filter_chain); + struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q); void tcf_block_put(struct tcf_block *block); int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp, struct tcf_result *res, bool compat_mode); @@ -30,7 +30,7 @@ int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp, #else static inline int tcf_block_get(struct tcf_block **p_block, - struct tcf_proto __rcu **p_filter_chain) + struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q) { return 0; } diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 684d8ed27eaa..df4032ca1b7f 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -270,6 +270,7 @@ struct tcf_chain { struct tcf_block { struct list_head chain_list; + struct Qdisc *q; }; static inline void qdisc_cb_private_validate(const struct sk_buff *skb, int sz) diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 2977b8a90851..f7d3f1f539b7 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -241,7 +241,7 @@ tcf_chain_filter_chain_ptr_set(struct tcf_chain *chain, } int tcf_block_get(struct tcf_block **p_block, - struct tcf_proto __rcu **p_filter_chain) + struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q) { struct tcf_block *block = kzalloc(sizeof(*block), GFP_KERNEL); struct tcf_chain *chain; @@ -257,6 +257,7 @@ int tcf_block_get(struct tcf_block **p_block, goto err_chain_create; } tcf_chain_filter_chain_ptr_set(chain, p_filter_chain); + block->q = q; *p_block = block; return 0; diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c index c5fcdf1a58a0..2dbd249c0b2f 100644 --- a/net/sched/sch_atm.c +++ b/net/sched/sch_atm.c @@ -281,7 +281,7 @@ static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent, goto err_out; } - error = tcf_block_get(&flow->block, &flow->filter_list); + error = tcf_block_get(&flow->block, &flow->filter_list, sch); if (error) { kfree(flow); goto err_out; @@ -546,7 +546,7 @@ static int atm_tc_init(struct Qdisc *sch, struct nlattr *opt) p->link.q = &noop_qdisc; pr_debug("atm_tc_init: link (%p) qdisc %p\n", &p->link, p->link.q); - err = tcf_block_get(&p->link.block, &p->link.filter_list); + err = tcf_block_get(&p->link.block, &p->link.filter_list, sch); if (err) return err; diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index dcef97fa8047..c3b92d62190e 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -1566,7 +1566,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct nlattr **t if (cl == NULL) goto failure; - err = tcf_block_get(&cl->block, &cl->filter_list); + err = tcf_block_get(&cl->block, &cl->filter_list, sch); if (err) { kfree(cl); return err; diff --git a/net/sched/sch_drr.c b/net/sched/sch_drr.c index 2d0e8d4bdc29..753dc7a77b60 100644 --- a/net/sched/sch_drr.c +++ b/net/sched/sch_drr.c @@ -412,7 +412,7 @@ static int drr_init_qdisc(struct Qdisc *sch, struct nlattr *opt) struct drr_sched *q = qdisc_priv(sch); int err; - err = tcf_block_get(&q->block, &q->filter_list); + err = tcf_block_get(&q->block, &q->filter_list, sch); if (err) return err; err = qdisc_class_hash_init(&q->clhash); diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c index 2836c80c7aa5..fb4fb71c68cf 100644 --- a/net/sched/sch_dsmark.c +++ b/net/sched/sch_dsmark.c @@ -344,7 +344,7 @@ static int dsmark_init(struct Qdisc *sch, struct nlattr *opt) if (!opt) goto errout; - err = tcf_block_get(&p->block, &p->filter_list); + err = tcf_block_get(&p->block, &p->filter_list, sch); if (err) return err; diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c index de3b57ceca7b..3c40edeff1e8 100644 --- a/net/sched/sch_fq_codel.c +++ b/net/sched/sch_fq_codel.c @@ -481,7 +481,7 @@ static int fq_codel_init(struct Qdisc *sch, struct nlattr *opt) return err; } - err = tcf_block_get(&q->block, &q->filter_list); + err = tcf_block_get(&q->block, &q->filter_list, sch); if (err) return err; diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index 3f88b75488b0..a692184bd333 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -1033,7 +1033,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid, if (cl == NULL) return -ENOBUFS; - err = tcf_block_get(&cl->block, &cl->filter_list); + err = tcf_block_get(&cl->block, &cl->filter_list, sch); if (err) { kfree(cl); return err; @@ -1405,7 +1405,7 @@ hfsc_init_qdisc(struct Qdisc *sch, struct nlattr *opt) return err; q->eligible = RB_ROOT; - err = tcf_block_get(&q->root.block, &q->root.filter_list); + err = tcf_block_get(&q->root.block, &q->root.filter_list, sch); if (err) return err; diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index c6d7ae81b41f..57be73c0e1d2 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -1030,7 +1030,7 @@ static int htb_init(struct Qdisc *sch, struct nlattr *opt) if (!opt) return -EINVAL; - err = tcf_block_get(&q->block, &q->filter_list); + err = tcf_block_get(&q->block, &q->filter_list, sch); if (err) return err; @@ -1393,7 +1393,7 @@ static int htb_change_class(struct Qdisc *sch, u32 classid, if (!cl) goto failure; - err = tcf_block_get(&cl->block, &cl->filter_list); + err = tcf_block_get(&cl->block, &cl->filter_list, sch); if (err) { kfree(cl); goto failure; diff --git a/net/sched/sch_ingress.c b/net/sched/sch_ingress.c index 44de4ee51ce9..9ccc1b89b0d9 100644 --- a/net/sched/sch_ingress.c +++ b/net/sched/sch_ingress.c @@ -59,7 +59,7 @@ static int ingress_init(struct Qdisc *sch, struct nlattr *opt) struct net_device *dev = qdisc_dev(sch); int err; - err = tcf_block_get(&q->block, &dev->ingress_cl_list); + err = tcf_block_get(&q->block, &dev->ingress_cl_list, sch); if (err) return err; @@ -153,11 +153,11 @@ static int clsact_init(struct Qdisc *sch, struct nlattr *opt) struct net_device *dev = qdisc_dev(sch); int err; - err = tcf_block_get(&q->ingress_block, &dev->ingress_cl_list); + err = tcf_block_get(&q->ingress_block, &dev->ingress_cl_list, sch); if (err) return err; - err = tcf_block_get(&q->egress_block, &dev->egress_cl_list); + err = tcf_block_get(&q->egress_block, &dev->egress_cl_list, sch); if (err) return err; diff --git a/net/sched/sch_multiq.c b/net/sched/sch_multiq.c index ff4fc3e0facd..31e0a284eeff 100644 --- a/net/sched/sch_multiq.c +++ b/net/sched/sch_multiq.c @@ -245,7 +245,7 @@ static int multiq_init(struct Qdisc *sch, struct nlattr *opt) if (opt == NULL) return -EINVAL; - err = tcf_block_get(&q->block, &q->filter_list); + err = tcf_block_get(&q->block, &q->filter_list, sch); if (err) return err; diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c index 2dd6c68ae91e..95fad348c8d7 100644 --- a/net/sched/sch_prio.c +++ b/net/sched/sch_prio.c @@ -212,7 +212,7 @@ static int prio_init(struct Qdisc *sch, struct nlattr *opt) if (!opt) return -EINVAL; - err = tcf_block_get(&q->block, &q->filter_list); + err = tcf_block_get(&q->block, &q->filter_list, sch); if (err) return err; diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c index 6ddfd4991108..8694c7b6d2b1 100644 --- a/net/sched/sch_qfq.c +++ b/net/sched/sch_qfq.c @@ -1419,7 +1419,7 @@ static int qfq_init_qdisc(struct Qdisc *sch, struct nlattr *opt) int i, j, err; u32 max_cl_shift, maxbudg_shift, max_classes; - err = tcf_block_get(&q->block, &q->filter_list); + err = tcf_block_get(&q->block, &q->filter_list, sch); if (err) return err; diff --git a/net/sched/sch_sfb.c b/net/sched/sch_sfb.c index cc39e170b4aa..487d375f5a06 100644 --- a/net/sched/sch_sfb.c +++ b/net/sched/sch_sfb.c @@ -553,7 +553,7 @@ static int sfb_init(struct Qdisc *sch, struct nlattr *opt) struct sfb_sched_data *q = qdisc_priv(sch); int err; - err = tcf_block_get(&q->block, &q->filter_list); + err = tcf_block_get(&q->block, &q->filter_list, sch); if (err) return err; diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index 74ea863b8240..123a53af2506 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -725,7 +725,7 @@ static int sfq_init(struct Qdisc *sch, struct nlattr *opt) setup_deferrable_timer(&q->perturb_timer, sfq_perturbation, (unsigned long)sch); - err = tcf_block_get(&q->block, &q->filter_list); + err = tcf_block_get(&q->block, &q->filter_list, sch); if (err) return err; -- cgit v1.2.3 From 855319becbcffec6988a4e781a861b69a71c5b58 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Fri, 13 Oct 2017 14:00:58 +0200 Subject: net: sched: store net pointer in block and introduce qdisc_net helper Store net pointer in the block structure. Along the way, introduce qdisc_net helper which allows to easily obtain net pointer for qdisc instance. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/pkt_sched.h | 7 +++++++ include/net/sch_generic.h | 1 + net/sched/cls_api.c | 1 + 3 files changed, 9 insertions(+) (limited to 'include/net') diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index 259bc191ba59..2d234af15f3e 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -4,7 +4,9 @@ #include #include #include +#include #include +#include #include #define DEFAULT_TX_QUEUE_LEN 1000 @@ -146,4 +148,9 @@ static inline bool is_classid_clsact_egress(u32 classid) TC_H_MIN(classid) == TC_H_MIN(TC_H_MIN_EGRESS); } +static inline struct net *qdisc_net(struct Qdisc *q) +{ + return dev_net(q->dev_queue->dev); +} + #endif diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index df4032ca1b7f..9b2cb91dc0d9 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -270,6 +270,7 @@ struct tcf_chain { struct tcf_block { struct list_head chain_list; + struct net *net; struct Qdisc *q; }; diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index f7d3f1f539b7..856003caa3bb 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -257,6 +257,7 @@ int tcf_block_get(struct tcf_block **p_block, goto err_chain_create; } tcf_chain_filter_chain_ptr_set(chain, p_filter_chain); + block->net = qdisc_net(q); block->q = q; *p_block = block; return 0; -- cgit v1.2.3 From 44186460c85a0121562db7cfef132d63c869118f Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Fri, 13 Oct 2017 14:00:59 +0200 Subject: net: sched: introduce tcf_block_q and tcf_block_dev helpers These helpers allows to get a q and netdev pointers for given block easily. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/pkt_cls.h | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) (limited to 'include/net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index e6c9e1e4d711..7bed674ba29a 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -24,6 +24,17 @@ void tcf_chain_put(struct tcf_chain *chain); int tcf_block_get(struct tcf_block **p_block, struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q); void tcf_block_put(struct tcf_block *block); + +static inline struct Qdisc *tcf_block_q(struct tcf_block *block) +{ + return block->q; +} + +static inline struct net_device *tcf_block_dev(struct tcf_block *block) +{ + return tcf_block_q(block)->dev_queue->dev; +} + int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp, struct tcf_result *res, bool compat_mode); @@ -39,6 +50,16 @@ static inline void tcf_block_put(struct tcf_block *block) { } +static inline struct Qdisc *tcf_block_q(struct tcf_block *block) +{ + return NULL; +} + +static inline struct net_device *tcf_block_dev(struct tcf_block *block) +{ + return NULL; +} + static inline int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp, struct tcf_result *res, bool compat_mode) { -- cgit v1.2.3 From 34e3759cf86a3e75463e34c1bb9691777406a175 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Fri, 13 Oct 2017 14:01:00 +0200 Subject: net: sched: teach tcf_bind/unbind_filter to use block->q Whenever the block->q is set, it can be used instead of tp->q as it contains the same value. When it is not set, which can't happen now but it might happen with the follow-up shared blocks introduction, the class is not set in the result. That would lead to a class lookup instead of direct class pointer use for classful qdiscs. However, it is not planned to support classful qdisqs sharing filter blocks, so that may never happen. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/pkt_cls.h | 27 +++++++++++++++++---------- 1 file changed, 17 insertions(+), 10 deletions(-) (limited to 'include/net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index 7bed674ba29a..49a143e0fe65 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -74,36 +74,43 @@ __cls_set_class(unsigned long *clp, unsigned long cl) } static inline unsigned long -cls_set_class(struct tcf_proto *tp, unsigned long *clp, - unsigned long cl) +cls_set_class(struct Qdisc *q, unsigned long *clp, unsigned long cl) { unsigned long old_cl; - - tcf_tree_lock(tp); + + sch_tree_lock(q); old_cl = __cls_set_class(clp, cl); - tcf_tree_unlock(tp); - + sch_tree_unlock(q); return old_cl; } static inline void tcf_bind_filter(struct tcf_proto *tp, struct tcf_result *r, unsigned long base) { + struct Qdisc *q = tp->chain->block->q; unsigned long cl; - cl = tp->q->ops->cl_ops->bind_tcf(tp->q, base, r->classid); - cl = cls_set_class(tp, &r->class, cl); + /* Check q as it is not set for shared blocks. In that case, + * setting class is not supported. + */ + if (!q) + return; + cl = q->ops->cl_ops->bind_tcf(q, base, r->classid); + cl = cls_set_class(q, &r->class, cl); if (cl) - tp->q->ops->cl_ops->unbind_tcf(tp->q, cl); + q->ops->cl_ops->unbind_tcf(q, cl); } static inline void tcf_unbind_filter(struct tcf_proto *tp, struct tcf_result *r) { + struct Qdisc *q = tp->chain->block->q; unsigned long cl; + if (!q) + return; if ((cl = __cls_set_class(&r->class, 0)) != 0) - tp->q->ops->cl_ops->unbind_tcf(tp->q, cl); + q->ops->cl_ops->unbind_tcf(q, cl); } struct tcf_exts { -- cgit v1.2.3 From 74e3be6021d22df2ffcb691eae1affeb2bd0128e Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Fri, 13 Oct 2017 14:01:04 +0200 Subject: net: sched: use tcf_block_q helper to get q pointer for sch_tree_lock Use tcf_block_q helper to get q pointer to be used for direct call of sch_tree_lock/unlock instead of tcf_tree_lock/unlock. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/sch_generic.h | 3 --- net/sched/sch_api.c | 6 ++++-- 2 files changed, 4 insertions(+), 5 deletions(-) (limited to 'include/net') diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 9b2cb91dc0d9..0aea9e23e97a 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -359,9 +359,6 @@ static inline void sch_tree_unlock(const struct Qdisc *q) spin_unlock_bh(qdisc_root_sleeping_lock(q)); } -#define tcf_tree_lock(tp) sch_tree_lock((tp)->q) -#define tcf_tree_unlock(tp) sch_tree_unlock((tp)->q) - extern struct Qdisc noop_qdisc; extern struct Qdisc_ops noop_qdisc_ops; extern struct Qdisc_ops pfifo_fast_ops; diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index aa82116ed10c..a9ac912f1d67 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -1661,9 +1661,11 @@ static int tcf_node_bind(struct tcf_proto *tp, void *n, struct tcf_walker *arg) struct tcf_bind_args *a = (void *)arg; if (tp->ops->bind_class) { - tcf_tree_lock(tp); + struct Qdisc *q = tcf_block_q(tp->chain->block); + + sch_tree_lock(q); tp->ops->bind_class(n, a->classid, a->cl); - tcf_tree_unlock(tp); + sch_tree_unlock(q); } return 0; } -- cgit v1.2.3 From 0da4af00b2ed3dbe46788623a696c4169447eadc Mon Sep 17 00:00:00 2001 From: Wei Wang Date: Fri, 13 Oct 2017 15:08:07 -0700 Subject: ipv6: only update __use and lastusetime once per jiffy at most In order to not dirty the cacheline too often, we try to only update dst->__use and dst->lastusetime at most once per jiffy. As dst->lastusetime is only used by ipv6 garbage collector, it should be good enough time resolution. And __use is only used in ipv6_route_seq_show() to show how many times a dst has been used. And as __use is not atomic_t right now, it does not show the precise number of usage times anyway. So we think it should be OK to only update it at most once per jiffy. According to my latest syn flood test on a machine with intel Xeon 6th gen processor and 2 10G mlx nics bonded together, each with 8 rx queues on 2 NUMA nodes: With this patch, the packet process rate increases from ~3.49Mpps to ~3.75Mpps with a 7% increase rate. Note: dst_use() is being renamed to dst_hold_and_use() to better specify the purpose of the function. Signed-off-by: Wei Wang Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/dst.h | 15 ++++++++------- net/decnet/dn_route.c | 8 ++++---- 2 files changed, 12 insertions(+), 11 deletions(-) (limited to 'include/net') diff --git a/include/net/dst.h b/include/net/dst.h index 204c19e25456..5047e8053d6c 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -255,17 +255,18 @@ static inline void dst_hold(struct dst_entry *dst) WARN_ON(atomic_inc_not_zero(&dst->__refcnt) == 0); } -static inline void dst_use(struct dst_entry *dst, unsigned long time) +static inline void dst_use_noref(struct dst_entry *dst, unsigned long time) { - dst_hold(dst); - dst->__use++; - dst->lastuse = time; + if (time != dst->lastuse) { + dst->__use++; + dst->lastuse = time; + } } -static inline void dst_use_noref(struct dst_entry *dst, unsigned long time) +static inline void dst_hold_and_use(struct dst_entry *dst, unsigned long time) { - dst->__use++; - dst->lastuse = time; + dst_hold(dst); + dst_use_noref(dst, time); } static inline struct dst_entry *dst_clone(struct dst_entry *dst) diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index 0bd3afd01dd2..bff5ab88cdbb 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -338,7 +338,7 @@ static int dn_insert_route(struct dn_route *rt, unsigned int hash, struct dn_rou dn_rt_hash_table[hash].chain); rcu_assign_pointer(dn_rt_hash_table[hash].chain, rth); - dst_use(&rth->dst, now); + dst_hold_and_use(&rth->dst, now); spin_unlock_bh(&dn_rt_hash_table[hash].lock); dst_release_immediate(&rt->dst); @@ -351,7 +351,7 @@ static int dn_insert_route(struct dn_route *rt, unsigned int hash, struct dn_rou rcu_assign_pointer(rt->dst.dn_next, dn_rt_hash_table[hash].chain); rcu_assign_pointer(dn_rt_hash_table[hash].chain, rt); - dst_use(&rt->dst, now); + dst_hold_and_use(&rt->dst, now); spin_unlock_bh(&dn_rt_hash_table[hash].lock); *rp = rt; return 0; @@ -1258,7 +1258,7 @@ static int __dn_route_output_key(struct dst_entry **pprt, const struct flowidn * (flp->flowidn_mark == rt->fld.flowidn_mark) && dn_is_output_route(rt) && (rt->fld.flowidn_oif == flp->flowidn_oif)) { - dst_use(&rt->dst, jiffies); + dst_hold_and_use(&rt->dst, jiffies); rcu_read_unlock_bh(); *pprt = &rt->dst; return 0; @@ -1535,7 +1535,7 @@ static int dn_route_input(struct sk_buff *skb) (rt->fld.flowidn_oif == 0) && (rt->fld.flowidn_mark == skb->mark) && (rt->fld.flowidn_iif == cb->iif)) { - dst_use(&rt->dst, jiffies); + dst_hold_and_use(&rt->dst, jiffies); rcu_read_unlock(); skb_dst_set(skb, (struct dst_entry *)rt); return 0; -- cgit v1.2.3 From a68f4a27f55f1d54e35c270aff89383da4b1b656 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 18 Oct 2017 11:36:39 +0100 Subject: rxrpc: Support service upgrade from a kernel service Provide support for a kernel service to make use of the service upgrade facility. This involves: (1) Pass an upgrade request flag to rxrpc_kernel_begin_call(). (2) Make rxrpc_kernel_recv_data() return the call's current service ID so that the caller can detect service upgrade and see what the service was upgraded to. Signed-off-by: David Howells --- Documentation/networking/rxrpc.txt | 17 +++++++++++++++-- fs/afs/internal.h | 1 + fs/afs/rxrpc.c | 11 +++++++---- include/net/af_rxrpc.h | 5 +++-- net/rxrpc/af_rxrpc.c | 5 ++++- net/rxrpc/recvmsg.c | 5 ++++- 6 files changed, 34 insertions(+), 10 deletions(-) (limited to 'include/net') diff --git a/Documentation/networking/rxrpc.txt b/Documentation/networking/rxrpc.txt index 810620153a44..9fb61a6bc7cf 100644 --- a/Documentation/networking/rxrpc.txt +++ b/Documentation/networking/rxrpc.txt @@ -782,7 +782,9 @@ The kernel interface functions are as follows: struct key *key, unsigned long user_call_ID, s64 tx_total_len, - gfp_t gfp); + gfp_t gfp, + rxrpc_notify_rx_t notify_rx, + bool upgrade); This allocates the infrastructure to make a new RxRPC call and assigns call and connection numbers. The call will be made on the UDP port that @@ -803,6 +805,13 @@ The kernel interface functions are as follows: allows the kernel to encrypt directly to the packet buffers, thereby saving a copy. The value may not be less than -1. + notify_rx is a pointer to a function to be called when events such as + incoming data packets or remote aborts happen. + + upgrade should be set to true if a client operation should request that + the server upgrade the service to a better one. The resultant service ID + is returned by rxrpc_kernel_recv_data(). + If this function is successful, an opaque reference to the RxRPC call is returned. The caller now holds a reference on this and it must be properly ended. @@ -850,7 +859,8 @@ The kernel interface functions are as follows: size_t size, size_t *_offset, bool want_more, - u32 *_abort) + u32 *_abort, + u16 *_service) This is used to receive data from either the reply part of a client call or the request part of a service call. buf and size specify how much @@ -873,6 +883,9 @@ The kernel interface functions are as follows: If a remote ABORT is detected, the abort code received will be stored in *_abort and ECONNABORTED will be returned. + The service ID that the call ended up with is returned into *_service. + This can be used to see if a call got a service upgrade. + (*) Abort a call. void rxrpc_kernel_abort_call(struct socket *sock, diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 82e16556afea..3f03f7888302 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -100,6 +100,7 @@ struct afs_call { bool send_pages; /* T if data from mapping should be sent */ bool need_attention; /* T if RxRPC poked us */ bool async; /* T if asynchronous */ + bool upgrade; /* T to request service upgrade */ u16 service_id; /* RxRPC service ID to call */ __be16 port; /* target UDP port */ u32 operation_ID; /* operation ID for an incoming call */ diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c index 0bf191f0dbaf..172a4f9747ac 100644 --- a/fs/afs/rxrpc.c +++ b/fs/afs/rxrpc.c @@ -387,7 +387,8 @@ int afs_make_call(struct in_addr *addr, struct afs_call *call, gfp_t gfp, tx_total_len, gfp, (async ? afs_wake_up_async_call : - afs_wake_up_call_waiter)); + afs_wake_up_call_waiter), + call->upgrade); call->key = NULL; if (IS_ERR(rxcall)) { ret = PTR_ERR(rxcall); @@ -443,7 +444,7 @@ error_do_abort: abort_code = 0; offset = 0; rxrpc_kernel_recv_data(afs_socket, rxcall, NULL, 0, &offset, - false, &abort_code); + false, &abort_code, &call->service_id); ret = call->type->abort_to_error(abort_code); } error_kill_call: @@ -471,7 +472,8 @@ static void afs_deliver_to_call(struct afs_call *call) size_t offset = 0; ret = rxrpc_kernel_recv_data(afs_socket, call->rxcall, NULL, 0, &offset, false, - &call->abort_code); + &call->abort_code, + &call->service_id); trace_afs_recv_data(call, 0, offset, false, ret); if (ret == -EINPROGRESS || ret == -EAGAIN) @@ -851,7 +853,8 @@ int afs_extract_data(struct afs_call *call, void *buf, size_t count, ret = rxrpc_kernel_recv_data(afs_socket, call->rxcall, buf, count, &call->offset, - want_more, &call->abort_code); + want_more, &call->abort_code, + &call->service_id); trace_afs_recv_data(call, count, call->offset, want_more, ret); if (ret == 0 || ret == -EAGAIN) return ret; diff --git a/include/net/af_rxrpc.h b/include/net/af_rxrpc.h index 3ac79150291f..820dd365a08e 100644 --- a/include/net/af_rxrpc.h +++ b/include/net/af_rxrpc.h @@ -49,12 +49,13 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *, unsigned long, s64, gfp_t, - rxrpc_notify_rx_t); + rxrpc_notify_rx_t, + bool); int rxrpc_kernel_send_data(struct socket *, struct rxrpc_call *, struct msghdr *, size_t, rxrpc_notify_end_tx_t); int rxrpc_kernel_recv_data(struct socket *, struct rxrpc_call *, - void *, size_t, size_t *, bool, u32 *); + void *, size_t, size_t *, bool, u32 *, u16 *); bool rxrpc_kernel_abort_call(struct socket *, struct rxrpc_call *, u32, int, const char *); void rxrpc_kernel_end_call(struct socket *, struct rxrpc_call *); diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index fb17552fd292..481f7dc90ba2 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -265,6 +265,7 @@ static int rxrpc_listen(struct socket *sock, int backlog) * @tx_total_len: Total length of data to transmit during the call (or -1) * @gfp: The allocation constraints * @notify_rx: Where to send notifications instead of socket queue + * @upgrade: Request service upgrade for call * * Allow a kernel service to begin a call on the nominated socket. This just * sets up all the internal tracking structures and allocates connection and @@ -279,7 +280,8 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, unsigned long user_call_ID, s64 tx_total_len, gfp_t gfp, - rxrpc_notify_rx_t notify_rx) + rxrpc_notify_rx_t notify_rx, + bool upgrade) { struct rxrpc_conn_parameters cp; struct rxrpc_call *call; @@ -304,6 +306,7 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, cp.key = key; cp.security_level = 0; cp.exclusive = false; + cp.upgrade = upgrade; cp.service_id = srx->srx_service; call = rxrpc_new_client_call(rx, &cp, srx, user_call_ID, tx_total_len, gfp); diff --git a/net/rxrpc/recvmsg.c b/net/rxrpc/recvmsg.c index bdece21f313d..e4937b3f3685 100644 --- a/net/rxrpc/recvmsg.c +++ b/net/rxrpc/recvmsg.c @@ -607,6 +607,7 @@ wait_error: * @_offset: The running offset into the buffer. * @want_more: True if more data is expected to be read * @_abort: Where the abort code is stored if -ECONNABORTED is returned + * @_service: Where to store the actual service ID (may be upgraded) * * Allow a kernel service to receive data and pick up information about the * state of a call. Returns 0 if got what was asked for and there's more @@ -624,7 +625,7 @@ wait_error: */ int rxrpc_kernel_recv_data(struct socket *sock, struct rxrpc_call *call, void *buf, size_t size, size_t *_offset, - bool want_more, u32 *_abort) + bool want_more, u32 *_abort, u16 *_service) { struct iov_iter iter; struct kvec iov; @@ -680,6 +681,8 @@ int rxrpc_kernel_recv_data(struct socket *sock, struct rxrpc_call *call, read_phase_complete: ret = 1; out: + if (_service) + *_service = call->service_id; mutex_unlock(&call->user_mutex); _leave(" = %d [%zu,%d]", ret, *_offset, *_abort); return ret; -- cgit v1.2.3 From f4d15fb6f99af9b99f688bd87579137be44f85ee Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 18 Oct 2017 11:07:31 +0100 Subject: rxrpc: Provide functions for allowing cleaner handling of signals Provide a couple of functions to allow cleaner handling of signals in a kernel service. They are: (1) rxrpc_kernel_get_rtt() This allows the kernel service to find out the RTT time for a call, so as to better judge how large a timeout to employ. Note, though, that whilst this returns a value in nanoseconds, the timeouts can only actually be in jiffies. (2) rxrpc_kernel_check_life() This returns a number that is updated when ACKs are received from the peer (notably including PING RESPONSE ACKs which we can elicit by sending PING ACKs to see if the call still exists on the server). The caller should compare the numbers of two calls to see if the call is still alive. These can be used to provide an extending timeout rather than returning immediately in the case that a signal occurs that would otherwise abort an RPC operation. The timeout would be extended if the server is still responsive and the call is still apparently alive on the server. For most operations this isn't that necessary - but for FS.StoreData it is: OpenAFS writes the data to storage as it comes in without making a backup, so if we immediately abort it when partially complete on a CTRL+C, say, we have no idea of the state of the file after the abort. Signed-off-by: David Howells --- Documentation/networking/rxrpc.txt | 24 ++++++++++++++++++++++++ include/net/af_rxrpc.h | 2 ++ net/rxrpc/af_rxrpc.c | 19 +++++++++++++++++++ net/rxrpc/peer_object.c | 13 +++++++++++++ 4 files changed, 58 insertions(+) (limited to 'include/net') diff --git a/Documentation/networking/rxrpc.txt b/Documentation/networking/rxrpc.txt index 9fb61a6bc7cf..1fb5c553aedd 100644 --- a/Documentation/networking/rxrpc.txt +++ b/Documentation/networking/rxrpc.txt @@ -1033,6 +1033,30 @@ The kernel interface functions are as follows: It returns 0 if the call was requeued and an error otherwise. + (*) Get call RTT. + + u64 rxrpc_kernel_get_rtt(struct socket *sock, struct rxrpc_call *call); + + Get the RTT time to the peer in use by a call. The value returned is in + nanoseconds. + + (*) Check call still alive. + + u32 rxrpc_kernel_check_life(struct socket *sock, + struct rxrpc_call *call); + + This returns a number that is updated when ACKs are received from the peer + (notably including PING RESPONSE ACKs which we can elicit by sending PING + ACKs to see if the call still exists on the server). The caller should + compare the numbers of two calls to see if the call is still alive after + waiting for a suitable interval. + + This allows the caller to work out if the server is still contactable and + if the call is still alive on the server whilst waiting for the server to + process a client operation. + + This function may transmit a PING ACK. + ======================= CONFIGURABLE PARAMETERS diff --git a/include/net/af_rxrpc.h b/include/net/af_rxrpc.h index 820dd365a08e..2b3a6eec4570 100644 --- a/include/net/af_rxrpc.h +++ b/include/net/af_rxrpc.h @@ -61,6 +61,7 @@ bool rxrpc_kernel_abort_call(struct socket *, struct rxrpc_call *, void rxrpc_kernel_end_call(struct socket *, struct rxrpc_call *); void rxrpc_kernel_get_peer(struct socket *, struct rxrpc_call *, struct sockaddr_rxrpc *); +u64 rxrpc_kernel_get_rtt(struct socket *, struct rxrpc_call *); int rxrpc_kernel_charge_accept(struct socket *, rxrpc_notify_rx_t, rxrpc_user_attach_call_t, unsigned long, gfp_t); void rxrpc_kernel_set_tx_length(struct socket *, struct rxrpc_call *, s64); @@ -68,5 +69,6 @@ int rxrpc_kernel_retry_call(struct socket *, struct rxrpc_call *, struct sockaddr_rxrpc *, struct key *); int rxrpc_kernel_check_call(struct socket *, struct rxrpc_call *, enum rxrpc_call_completion *, u32 *); +u32 rxrpc_kernel_check_life(struct socket *, struct rxrpc_call *); #endif /* _NET_RXRPC_H */ diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 481f7dc90ba2..73c980e26581 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -339,6 +339,25 @@ void rxrpc_kernel_end_call(struct socket *sock, struct rxrpc_call *call) } EXPORT_SYMBOL(rxrpc_kernel_end_call); +/** + * rxrpc_kernel_check_life - Check to see whether a call is still alive + * @sock: The socket the call is on + * @call: The call to check + * + * Allow a kernel service to find out whether a call is still alive - ie. we're + * getting ACKs from the server. Returns a number representing the life state + * which can be compared to that returned by a previous call. + * + * If this is a client call, ping ACKs will be sent to the server to find out + * whether it's still responsive and whether the call is still alive on the + * server. + */ +u32 rxrpc_kernel_check_life(struct socket *sock, struct rxrpc_call *call) +{ + return call->acks_latest; +} +EXPORT_SYMBOL(rxrpc_kernel_check_life); + /** * rxrpc_kernel_check_call - Check a call's state * @sock: The socket the call is on diff --git a/net/rxrpc/peer_object.c b/net/rxrpc/peer_object.c index 5787f97f5330..d02a99f37f5f 100644 --- a/net/rxrpc/peer_object.c +++ b/net/rxrpc/peer_object.c @@ -411,3 +411,16 @@ void rxrpc_kernel_get_peer(struct socket *sock, struct rxrpc_call *call, *_srx = call->peer->srx; } EXPORT_SYMBOL(rxrpc_kernel_get_peer); + +/** + * rxrpc_kernel_get_rtt - Get a call's peer RTT + * @sock: The socket on which the call is in progress. + * @call: The call to query + * + * Get the call's peer RTT. + */ +u64 rxrpc_kernel_get_rtt(struct socket *sock, struct rxrpc_call *call) +{ + return call->peer->rtt; +} +EXPORT_SYMBOL(rxrpc_kernel_get_rtt); -- cgit v1.2.3 From f8b8b1cd5aadd221742b45eb0ee3c8a80abf036a Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 16 Oct 2017 11:12:18 -0400 Subject: net: dsa: split dsa_port's netdev member The dsa_port structure has a "netdev" member, which can be used for either the master device, or the slave device, depending on its type. It is true that today, CPU port are not exposed to userspace, thus the port's netdev member can be used to point to its master interface. But it is still slightly confusing, so split it into more explicit "master" and "slave" members inside an anonymous union. Signed-off-by: Vivien Didelot Signed-off-by: David S. Miller --- drivers/net/dsa/bcm_sf2.c | 6 +++--- drivers/net/dsa/mt7530.c | 2 +- drivers/net/dsa/mv88e6xxx/chip.c | 2 +- include/net/dsa.h | 9 ++++++++- net/dsa/dsa.c | 6 +++--- net/dsa/dsa2.c | 20 ++++++++++---------- net/dsa/dsa_priv.h | 4 ++-- net/dsa/legacy.c | 14 +++++++------- net/dsa/slave.c | 6 +++--- 9 files changed, 38 insertions(+), 31 deletions(-) (limited to 'include/net') diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c index 32025b990437..b43c063b9634 100644 --- a/drivers/net/dsa/bcm_sf2.c +++ b/drivers/net/dsa/bcm_sf2.c @@ -601,7 +601,7 @@ static void bcm_sf2_sw_fixed_link_update(struct dsa_switch *ds, int port, * state machine and make it go in PHY_FORCING state instead. */ if (!status->link) - netif_carrier_off(ds->ports[port].netdev); + netif_carrier_off(ds->ports[port].slave); status->duplex = 1; } else { status->link = 1; @@ -690,7 +690,7 @@ static int bcm_sf2_sw_resume(struct dsa_switch *ds) static void bcm_sf2_sw_get_wol(struct dsa_switch *ds, int port, struct ethtool_wolinfo *wol) { - struct net_device *p = ds->ports[port].cpu_dp->netdev; + struct net_device *p = ds->ports[port].cpu_dp->master; struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds); struct ethtool_wolinfo pwol; @@ -713,7 +713,7 @@ static void bcm_sf2_sw_get_wol(struct dsa_switch *ds, int port, static int bcm_sf2_sw_set_wol(struct dsa_switch *ds, int port, struct ethtool_wolinfo *wol) { - struct net_device *p = ds->ports[port].cpu_dp->netdev; + struct net_device *p = ds->ports[port].cpu_dp->master; struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds); s8 cpu_port = ds->ports[port].cpu_dp->index; struct ethtool_wolinfo pwol; diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c index 034241696ce2..fea2e665d0cb 100644 --- a/drivers/net/dsa/mt7530.c +++ b/drivers/net/dsa/mt7530.c @@ -933,7 +933,7 @@ mt7530_setup(struct dsa_switch *ds) * controller also is the container for two GMACs nodes representing * as two netdev instances. */ - dn = ds->ports[MT7530_CPU_PORT].netdev->dev.of_node->parent; + dn = ds->ports[MT7530_CPU_PORT].master->dev.of_node->parent; priv->ethernet = syscon_node_to_regmap(dn); if (IS_ERR(priv->ethernet)) return PTR_ERR(priv->ethernet); diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index 76cf383dcf90..88b47fa09b41 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -1137,7 +1137,7 @@ static int mv88e6xxx_port_check_hw_vlan(struct dsa_switch *ds, int port, if (dsa_is_dsa_port(ds, i) || dsa_is_cpu_port(ds, i)) continue; - if (!ds->ports[port].netdev) + if (!ds->ports[port].slave) continue; if (vlan.member[i] == diff --git a/include/net/dsa.h b/include/net/dsa.h index 2746741f74cf..6ed1a17ed1bd 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -164,6 +164,14 @@ struct dsa_mall_tc_entry { struct dsa_port { + /* A CPU port is physically connected to a master device. + * A user port exposed to userspace has a slave device. + */ + union { + struct net_device *master; + struct net_device *slave; + }; + /* CPU port tagging operations used by master or slave devices */ const struct dsa_device_ops *tag_ops; @@ -176,7 +184,6 @@ struct dsa_port { unsigned int index; const char *name; struct dsa_port *cpu_dp; - struct net_device *netdev; struct device_node *dn; unsigned int ageing_time; u8 stp_state; diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index 832c659ff993..a3abf7a7b9a2 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -201,7 +201,7 @@ static int dsa_switch_rcv(struct sk_buff *skb, struct net_device *dev, #ifdef CONFIG_PM_SLEEP static bool dsa_is_port_initialized(struct dsa_switch *ds, int p) { - return ds->enabled_port_mask & (1 << p) && ds->ports[p].netdev; + return ds->enabled_port_mask & (1 << p) && ds->ports[p].slave; } int dsa_switch_suspend(struct dsa_switch *ds) @@ -213,7 +213,7 @@ int dsa_switch_suspend(struct dsa_switch *ds) if (!dsa_is_port_initialized(ds, i)) continue; - ret = dsa_slave_suspend(ds->ports[i].netdev); + ret = dsa_slave_suspend(ds->ports[i].slave); if (ret) return ret; } @@ -240,7 +240,7 @@ int dsa_switch_resume(struct dsa_switch *ds) if (!dsa_is_port_initialized(ds, i)) continue; - ret = dsa_slave_resume(ds->ports[i].netdev); + ret = dsa_slave_resume(ds->ports[i].slave); if (ret) return ret; } diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index 6ac9e11d385c..9e8b8aab049d 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -279,7 +279,7 @@ static int dsa_user_port_apply(struct dsa_port *port) if (err) { dev_warn(ds->dev, "Failed to create slave %d: %d\n", port->index, err); - port->netdev = NULL; + port->slave = NULL; return err; } @@ -289,7 +289,7 @@ static int dsa_user_port_apply(struct dsa_port *port) if (err) return err; - devlink_port_type_eth_set(&port->devlink_port, port->netdev); + devlink_port_type_eth_set(&port->devlink_port, port->slave); return 0; } @@ -297,9 +297,9 @@ static int dsa_user_port_apply(struct dsa_port *port) static void dsa_user_port_unapply(struct dsa_port *port) { devlink_port_unregister(&port->devlink_port); - if (port->netdev) { - dsa_slave_destroy(port->netdev); - port->netdev = NULL; + if (port->slave) { + dsa_slave_destroy(port->slave); + port->slave = NULL; port->ds->enabled_port_mask &= ~(1 << port->index); } } @@ -432,9 +432,9 @@ static int dsa_dst_apply(struct dsa_switch_tree *dst) * sent to the tag format's receive function. */ wmb(); - dst->cpu_dp->netdev->dsa_ptr = dst->cpu_dp; + dst->cpu_dp->master->dsa_ptr = dst->cpu_dp; - err = dsa_master_ethtool_setup(dst->cpu_dp->netdev); + err = dsa_master_ethtool_setup(dst->cpu_dp->master); if (err) return err; @@ -451,9 +451,9 @@ static void dsa_dst_unapply(struct dsa_switch_tree *dst) if (!dst->applied) return; - dsa_master_ethtool_restore(dst->cpu_dp->netdev); + dsa_master_ethtool_restore(dst->cpu_dp->master); - dst->cpu_dp->netdev->dsa_ptr = NULL; + dst->cpu_dp->master->dsa_ptr = NULL; /* If we used a tagging format that doesn't have an ethertype * field, make sure that all packets from this point get sent @@ -499,7 +499,7 @@ static int dsa_cpu_parse(struct dsa_port *port, u32 index, if (!dst->cpu_dp) { dst->cpu_dp = port; - dst->cpu_dp->netdev = ethernet_dev; + dst->cpu_dp->master = ethernet_dev; } /* Initialize cpu_port_mask now for drv->setup() diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h index 623c22b75e81..1e9914062d0b 100644 --- a/net/dsa/dsa_priv.h +++ b/net/dsa/dsa_priv.h @@ -130,7 +130,7 @@ static inline struct net_device *dsa_master_find_slave(struct net_device *dev, if (port < 0 || port >= ds->num_ports) return NULL; - return ds->ports[port].netdev; + return ds->ports[port].slave; } /* port.c */ @@ -181,7 +181,7 @@ dsa_slave_to_master(const struct net_device *dev) { struct dsa_port *dp = dsa_slave_to_port(dev); - return dp->cpu_dp->netdev; + return dp->cpu_dp->master; } /* switch.c */ diff --git a/net/dsa/legacy.c b/net/dsa/legacy.c index cc28c6f792a3..b6c88fd33d4f 100644 --- a/net/dsa/legacy.c +++ b/net/dsa/legacy.c @@ -120,7 +120,7 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, return -EINVAL; } dst->cpu_dp = &ds->ports[i]; - dst->cpu_dp->netdev = master; + dst->cpu_dp->master = master; ds->cpu_port_mask |= 1 << i; } else if (!strcmp(name, "dsa")) { ds->dsa_port_mask |= 1 << i; @@ -261,10 +261,10 @@ static void dsa_switch_destroy(struct dsa_switch *ds) if (!(ds->enabled_port_mask & (1 << port))) continue; - if (!ds->ports[port].netdev) + if (!ds->ports[port].slave) continue; - dsa_slave_destroy(ds->ports[port].netdev); + dsa_slave_destroy(ds->ports[port].slave); } /* Disable configuration of the CPU and DSA ports */ @@ -601,7 +601,7 @@ static int dsa_setup_dst(struct dsa_switch_tree *dst, struct net_device *dev, wmb(); dev->dsa_ptr = dst->cpu_dp; - return dsa_master_ethtool_setup(dst->cpu_dp->netdev); + return dsa_master_ethtool_setup(dst->cpu_dp->master); } static int dsa_probe(struct platform_device *pdev) @@ -666,9 +666,9 @@ static void dsa_remove_dst(struct dsa_switch_tree *dst) { int i; - dsa_master_ethtool_restore(dst->cpu_dp->netdev); + dsa_master_ethtool_restore(dst->cpu_dp->master); - dst->cpu_dp->netdev->dsa_ptr = NULL; + dst->cpu_dp->master->dsa_ptr = NULL; /* If we used a tagging format that doesn't have an ethertype * field, make sure that all packets from this point get sent @@ -683,7 +683,7 @@ static void dsa_remove_dst(struct dsa_switch_tree *dst) dsa_switch_destroy(ds); } - dev_put(dst->cpu_dp->netdev); + dev_put(dst->cpu_dp->master); } static int dsa_remove(struct platform_device *pdev) diff --git a/net/dsa/slave.c b/net/dsa/slave.c index b72e07503a40..6906de0f0050 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -1123,7 +1123,7 @@ static void dsa_slave_notify(struct net_device *dev, unsigned long val) int dsa_slave_create(struct dsa_port *port, const char *name) { struct dsa_port *cpu_dp = port->cpu_dp; - struct net_device *master = cpu_dp->netdev; + struct net_device *master = cpu_dp->master; struct dsa_switch *ds = port->ds; struct net_device *slave_dev; struct dsa_slave_priv *p; @@ -1170,7 +1170,7 @@ int dsa_slave_create(struct dsa_port *port, const char *name) p->old_link = -1; p->old_duplex = -1; - port->netdev = slave_dev; + port->slave = slave_dev; netif_carrier_off(slave_dev); @@ -1198,7 +1198,7 @@ out_phy: out_free: free_percpu(p->stats64); free_netdev(slave_dev); - port->netdev = NULL; + port->slave = NULL; return ret; } -- cgit v1.2.3 From c8652c83bc84ac8db44060ced0036de7628aa5e5 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 16 Oct 2017 11:12:19 -0400 Subject: net: dsa: add dsa_to_port helper The dsa_port structure is part of DSA core data and must only be updated by the later. It is OK and sometimes necessary for the DSA drivers to access this data, but this has to be read only. For that purpose, add a dsa_to_port() helper which returns a const pointer to a dsa_port structure which must be used by DSA drivers from now on instead of digging into ds->ports[] themselves. Signed-off-by: Vivien Didelot Signed-off-by: David S. Miller --- drivers/net/dsa/b53/b53_common.c | 4 ++-- drivers/net/dsa/lan9303-core.c | 2 +- drivers/net/dsa/mt7530.c | 4 ++-- drivers/net/dsa/mv88e6060.c | 2 +- drivers/net/dsa/mv88e6xxx/chip.c | 8 ++++---- drivers/net/dsa/qca8k.c | 4 ++-- include/net/dsa.h | 5 +++++ 7 files changed, 17 insertions(+), 12 deletions(-) (limited to 'include/net') diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index d4ce092def83..b48cf0487b43 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -1354,7 +1354,7 @@ int b53_br_join(struct dsa_switch *ds, int port, struct net_device *br) b53_read16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(port), &pvlan); b53_for_each_port(dev, i) { - if (ds->ports[i].bridge_dev != br) + if (dsa_to_port(ds, i)->bridge_dev != br) continue; /* Add this local port to the remote port VLAN control @@ -1390,7 +1390,7 @@ void b53_br_leave(struct dsa_switch *ds, int port, struct net_device *br) b53_for_each_port(dev, i) { /* Don't touch the remaining ports */ - if (ds->ports[i].bridge_dev != br) + if (dsa_to_port(ds, i)->bridge_dev != br) continue; b53_read16(dev, B53_PVLAN_PAGE, B53_PVLAN_PORT_MASK(i), ®); diff --git a/drivers/net/dsa/lan9303-core.c b/drivers/net/dsa/lan9303-core.c index fecfe1fe67ea..09a748327fc6 100644 --- a/drivers/net/dsa/lan9303-core.c +++ b/drivers/net/dsa/lan9303-core.c @@ -863,7 +863,7 @@ static int lan9303_port_bridge_join(struct dsa_switch *ds, int port, struct lan9303 *chip = ds->priv; dev_dbg(chip->dev, "%s(port %d)\n", __func__, port); - if (ds->ports[1].bridge_dev == ds->ports[2].bridge_dev) { + if (dsa_to_port(ds, 1)->bridge_dev == dsa_to_port(ds, 2)->bridge_dev) { lan9303_bridge_ports(chip); chip->is_bridged = true; /* unleash stp_state_set() */ } diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c index fea2e665d0cb..21431be2831e 100644 --- a/drivers/net/dsa/mt7530.c +++ b/drivers/net/dsa/mt7530.c @@ -782,7 +782,7 @@ mt7530_port_bridge_join(struct dsa_switch *ds, int port, * and not being setup until the port becomes enabled. */ if (ds->enabled_port_mask & BIT(i) && i != port) { - if (ds->ports[i].bridge_dev != bridge) + if (dsa_to_port(ds, i)->bridge_dev != bridge) continue; if (priv->ports[i].enable) mt7530_set(priv, MT7530_PCR_P(i), @@ -819,7 +819,7 @@ mt7530_port_bridge_leave(struct dsa_switch *ds, int port, * is kept and not being setup until the port becomes enabled. */ if (ds->enabled_port_mask & BIT(i) && i != port) { - if (ds->ports[i].bridge_dev != bridge) + if (dsa_to_port(ds, i)->bridge_dev != bridge) continue; if (priv->ports[i].enable) mt7530_clear(priv, MT7530_PCR_P(i), diff --git a/drivers/net/dsa/mv88e6060.c b/drivers/net/dsa/mv88e6060.c index 6173be889d95..f78b9e13be1c 100644 --- a/drivers/net/dsa/mv88e6060.c +++ b/drivers/net/dsa/mv88e6060.c @@ -177,7 +177,7 @@ static int mv88e6060_setup_port(struct dsa_switch *ds, int p) ((p & 0xf) << PORT_VLAN_MAP_DBNUM_SHIFT) | (dsa_is_cpu_port(ds, p) ? ds->enabled_port_mask : - BIT(ds->ports[p].cpu_dp->index))); + BIT(dsa_to_port(ds, p)->cpu_dp->index))); /* Port Association Vector: when learning source addresses * of packets, add the address to the address database using diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index 88b47fa09b41..677d6902807e 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -851,7 +851,7 @@ static u16 mv88e6xxx_port_vlan(struct mv88e6xxx_chip *chip, int dev, int port) for (i = 0; i < mv88e6xxx_num_ports(chip); ++i) if (dsa_is_cpu_port(chip->ds, i) || dsa_is_dsa_port(chip->ds, i) || - (br && chip->ds->ports[i].bridge_dev == br)) + (br && dsa_to_port(chip->ds, i)->bridge_dev == br)) pvlan |= BIT(i); return pvlan; @@ -1144,16 +1144,16 @@ static int mv88e6xxx_port_check_hw_vlan(struct dsa_switch *ds, int port, MV88E6XXX_G1_VTU_DATA_MEMBER_TAG_NON_MEMBER) continue; - if (ds->ports[i].bridge_dev == + if (dsa_to_port(ds, i)->bridge_dev == ds->ports[port].bridge_dev) break; /* same bridge, check next VLAN */ - if (!ds->ports[i].bridge_dev) + if (!dsa_to_port(ds, i)->bridge_dev) continue; dev_err(ds->dev, "p%d: hw VLAN %d already used by %s\n", port, vlan.vid, - netdev_name(ds->ports[i].bridge_dev)); + netdev_name(dsa_to_port(ds, i)->bridge_dev)); err = -EOPNOTSUPP; goto unlock; } diff --git a/drivers/net/dsa/qca8k.c b/drivers/net/dsa/qca8k.c index 82f09711ac1a..d1b0b1fb632f 100644 --- a/drivers/net/dsa/qca8k.c +++ b/drivers/net/dsa/qca8k.c @@ -700,7 +700,7 @@ qca8k_port_bridge_join(struct dsa_switch *ds, int port, struct net_device *br) int i; for (i = 1; i < QCA8K_NUM_PORTS; i++) { - if (ds->ports[i].bridge_dev != br) + if (dsa_to_port(ds, i)->bridge_dev != br) continue; /* Add this port to the portvlan mask of the other ports * in the bridge @@ -725,7 +725,7 @@ qca8k_port_bridge_leave(struct dsa_switch *ds, int port, struct net_device *br) int i; for (i = 1; i < QCA8K_NUM_PORTS; i++) { - if (ds->ports[i].bridge_dev != br) + if (dsa_to_port(ds, i)->bridge_dev != br) continue; /* Remove this port to the portvlan mask of the other ports * in the bridge diff --git a/include/net/dsa.h b/include/net/dsa.h index 6ed1a17ed1bd..38961ef91d3d 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -269,6 +269,11 @@ static inline bool dsa_is_normal_port(struct dsa_switch *ds, int p) return !dsa_is_cpu_port(ds, p) && !dsa_is_dsa_port(ds, p); } +static inline const struct dsa_port *dsa_to_port(struct dsa_switch *ds, int p) +{ + return &ds->ports[p]; +} + static inline u8 dsa_upstream_port(struct dsa_switch *ds) { struct dsa_switch_tree *dst = ds->dst; -- cgit v1.2.3 From eb4ddaf474285a4c6986f4a1c3205bdb0bed2da9 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 16 Oct 2017 17:28:45 -0700 Subject: net/decnet: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: "David S. Miller" Cc: Johannes Berg Cc: David Ahern Cc: linux-decnet-user@lists.sourceforge.net Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook Signed-off-by: David S. Miller --- include/net/dn.h | 7 ------- include/net/dn_nsp.h | 1 - net/decnet/af_decnet.c | 4 ---- net/decnet/dn_dev.c | 12 +++++------- net/decnet/dn_nsp_out.c | 11 ----------- 5 files changed, 5 insertions(+), 30 deletions(-) (limited to 'include/net') diff --git a/include/net/dn.h b/include/net/dn.h index 913b73d239f5..4394f7d5cfe8 100644 --- a/include/net/dn.h +++ b/include/net/dn.h @@ -122,13 +122,6 @@ struct dn_scp /* Session Control Port */ unsigned long keepalive; void (*keepalive_fxn)(struct sock *sk); - /* - * This stuff is for the fast timer for delayed acks - */ - struct timer_list delack_timer; - int delack_pending; - void (*delack_fxn)(struct sock *sk); - }; static inline struct dn_scp *DN_SK(struct sock *sk) diff --git a/include/net/dn_nsp.h b/include/net/dn_nsp.h index 3a3e33d18456..413a15e5339c 100644 --- a/include/net/dn_nsp.h +++ b/include/net/dn_nsp.h @@ -17,7 +17,6 @@ void dn_nsp_send_data_ack(struct sock *sk); void dn_nsp_send_oth_ack(struct sock *sk); -void dn_nsp_delayed_ack(struct sock *sk); void dn_send_conn_ack(struct sock *sk); void dn_send_conn_conf(struct sock *sk, gfp_t gfp); void dn_nsp_send_disc(struct sock *sk, unsigned char type, diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c index 73a0399dc7a2..d4c9a8bbad3e 100644 --- a/net/decnet/af_decnet.c +++ b/net/decnet/af_decnet.c @@ -533,10 +533,6 @@ static struct sock *dn_alloc_sock(struct net *net, struct socket *sock, gfp_t gf scp->keepalive = 10 * HZ; scp->keepalive_fxn = dn_keepalive; - init_timer(&scp->delack_timer); - scp->delack_pending = 0; - scp->delack_fxn = dn_nsp_delayed_ack; - dn_start_slow_timer(sk); out: return sk; diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c index 4d339de56862..92dbaa3f1eae 100644 --- a/net/decnet/dn_dev.c +++ b/net/decnet/dn_dev.c @@ -1038,14 +1038,14 @@ static void dn_eth_down(struct net_device *dev) static void dn_dev_set_timer(struct net_device *dev); -static void dn_dev_timer_func(unsigned long arg) +static void dn_dev_timer_func(struct timer_list *t) { - struct net_device *dev = (struct net_device *)arg; - struct dn_dev *dn_db; + struct dn_dev *dn_db = from_timer(dn_db, t, timer); + struct net_device *dev; struct dn_ifaddr *ifa; rcu_read_lock(); - dn_db = rcu_dereference(dev->dn_ptr); + dev = dn_db->dev; if (dn_db->t3 <= dn_db->parms.t2) { if (dn_db->parms.timer3) { for (ifa = rcu_dereference(dn_db->ifa_list); @@ -1070,8 +1070,6 @@ static void dn_dev_set_timer(struct net_device *dev) if (dn_db->parms.t2 > dn_db->parms.t3) dn_db->parms.t2 = dn_db->parms.t3; - dn_db->timer.data = (unsigned long)dev; - dn_db->timer.function = dn_dev_timer_func; dn_db->timer.expires = jiffies + (dn_db->parms.t2 * HZ); add_timer(&dn_db->timer); @@ -1100,7 +1098,7 @@ static struct dn_dev *dn_dev_create(struct net_device *dev, int *err) rcu_assign_pointer(dev->dn_ptr, dn_db); dn_db->dev = dev; - init_timer(&dn_db->timer); + timer_setup(&dn_db->timer, dn_dev_timer_func, 0); dn_db->uptime = jiffies; diff --git a/net/decnet/dn_nsp_out.c b/net/decnet/dn_nsp_out.c index 66f035e476ea..e50a4adfcf7e 100644 --- a/net/decnet/dn_nsp_out.c +++ b/net/decnet/dn_nsp_out.c @@ -491,17 +491,6 @@ void dn_send_conn_ack (struct sock *sk) dn_nsp_send(skb); } -void dn_nsp_delayed_ack(struct sock *sk) -{ - struct dn_scp *scp = DN_SK(sk); - - if (scp->ackxmt_oth != scp->numoth_rcv) - dn_nsp_send_oth_ack(sk); - - if (scp->ackxmt_dat != scp->numdat_rcv) - dn_nsp_send_data_ack(sk); -} - static int dn_nsp_retrans_conn_conf(struct sock *sk) { struct dn_scp *scp = DN_SK(sk); -- cgit v1.2.3 From 59f379f9046a9e0532ffd19b44e3c32fe79ec51b Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 16 Oct 2017 17:29:19 -0700 Subject: inet/connection_sock: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: "David S. Miller" Cc: Gerrit Renker Cc: Alexey Kuznetsov Cc: Hideaki YOSHIFUJI Cc: netdev@vger.kernel.org Cc: dccp@vger.kernel.org Signed-off-by: Kees Cook Signed-off-by: David S. Miller --- include/net/inet_connection_sock.h | 6 +++--- net/dccp/timer.c | 18 ++++++++++-------- net/ipv4/inet_connection_sock.c | 21 +++++++++------------ net/ipv4/tcp_timer.c | 18 +++++++++++------- 4 files changed, 33 insertions(+), 30 deletions(-) (limited to 'include/net') diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index 13e4c89a8231..0358745ea059 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -169,9 +169,9 @@ enum inet_csk_ack_state_t { }; void inet_csk_init_xmit_timers(struct sock *sk, - void (*retransmit_handler)(unsigned long), - void (*delack_handler)(unsigned long), - void (*keepalive_handler)(unsigned long)); + void (*retransmit_handler)(struct timer_list *), + void (*delack_handler)(struct timer_list *), + void (*keepalive_handler)(struct timer_list *)); void inet_csk_clear_xmit_timers(struct sock *sk); static inline void inet_csk_schedule_ack(struct sock *sk) diff --git a/net/dccp/timer.c b/net/dccp/timer.c index 3a2c34027758..1e35526bf436 100644 --- a/net/dccp/timer.c +++ b/net/dccp/timer.c @@ -125,10 +125,11 @@ static void dccp_retransmit_timer(struct sock *sk) __sk_dst_reset(sk); } -static void dccp_write_timer(unsigned long data) +static void dccp_write_timer(struct timer_list *t) { - struct sock *sk = (struct sock *)data; - struct inet_connection_sock *icsk = inet_csk(sk); + struct inet_connection_sock *icsk = + from_timer(icsk, t, icsk_retransmit_timer); + struct sock *sk = &icsk->icsk_inet.sk; int event = 0; bh_lock_sock(sk); @@ -161,19 +162,20 @@ out: sock_put(sk); } -static void dccp_keepalive_timer(unsigned long data) +static void dccp_keepalive_timer(struct timer_list *t) { - struct sock *sk = (struct sock *)data; + struct sock *sk = from_timer(sk, t, sk_timer); pr_err("dccp should not use a keepalive timer !\n"); sock_put(sk); } /* This is the same as tcp_delack_timer, sans prequeue & mem_reclaim stuff */ -static void dccp_delack_timer(unsigned long data) +static void dccp_delack_timer(struct timer_list *t) { - struct sock *sk = (struct sock *)data; - struct inet_connection_sock *icsk = inet_csk(sk); + struct inet_connection_sock *icsk = + from_timer(icsk, t, icsk_delack_timer); + struct sock *sk = &icsk->icsk_inet.sk; bh_lock_sock(sk); if (sock_owned_by_user(sk)) { diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 8a91ebbf0c01..5c965ecc96a0 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -494,17 +494,15 @@ EXPORT_SYMBOL(inet_csk_accept); * to optimize. */ void inet_csk_init_xmit_timers(struct sock *sk, - void (*retransmit_handler)(unsigned long), - void (*delack_handler)(unsigned long), - void (*keepalive_handler)(unsigned long)) + void (*retransmit_handler)(struct timer_list *t), + void (*delack_handler)(struct timer_list *t), + void (*keepalive_handler)(struct timer_list *t)) { struct inet_connection_sock *icsk = inet_csk(sk); - setup_timer(&icsk->icsk_retransmit_timer, retransmit_handler, - (unsigned long)sk); - setup_timer(&icsk->icsk_delack_timer, delack_handler, - (unsigned long)sk); - setup_timer(&sk->sk_timer, keepalive_handler, (unsigned long)sk); + timer_setup(&icsk->icsk_retransmit_timer, retransmit_handler, 0); + timer_setup(&icsk->icsk_delack_timer, delack_handler, 0); + timer_setup(&sk->sk_timer, keepalive_handler, 0); icsk->icsk_pending = icsk->icsk_ack.pending = 0; } EXPORT_SYMBOL(inet_csk_init_xmit_timers); @@ -676,9 +674,9 @@ void inet_csk_reqsk_queue_drop_and_put(struct sock *sk, struct request_sock *req } EXPORT_SYMBOL(inet_csk_reqsk_queue_drop_and_put); -static void reqsk_timer_handler(unsigned long data) +static void reqsk_timer_handler(struct timer_list *t) { - struct request_sock *req = (struct request_sock *)data; + struct request_sock *req = from_timer(req, t, rsk_timer); struct sock *sk_listener = req->rsk_listener; struct net *net = sock_net(sk_listener); struct inet_connection_sock *icsk = inet_csk(sk_listener); @@ -749,8 +747,7 @@ static void reqsk_queue_hash_req(struct request_sock *req, req->num_timeout = 0; req->sk = NULL; - setup_pinned_timer(&req->rsk_timer, reqsk_timer_handler, - (unsigned long)req); + timer_setup(&req->rsk_timer, reqsk_timer_handler, TIMER_PINNED); mod_timer(&req->rsk_timer, jiffies + timeout); inet_ehash_insert(req_to_sk(req), NULL); diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 7014cc00c74c..804a8d34ce86 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -288,15 +288,17 @@ out: * * Returns: Nothing (void) */ -static void tcp_delack_timer(unsigned long data) +static void tcp_delack_timer(struct timer_list *t) { - struct sock *sk = (struct sock *)data; + struct inet_connection_sock *icsk = + from_timer(icsk, t, icsk_delack_timer); + struct sock *sk = &icsk->icsk_inet.sk; bh_lock_sock(sk); if (!sock_owned_by_user(sk)) { tcp_delack_timer_handler(sk); } else { - inet_csk(sk)->icsk_ack.blocked = 1; + icsk->icsk_ack.blocked = 1; __NET_INC_STATS(sock_net(sk), LINUX_MIB_DELAYEDACKLOCKED); /* deleguate our work to tcp_release_cb() */ if (!test_and_set_bit(TCP_DELACK_TIMER_DEFERRED, &sk->sk_tsq_flags)) @@ -576,9 +578,11 @@ out: sk_mem_reclaim(sk); } -static void tcp_write_timer(unsigned long data) +static void tcp_write_timer(struct timer_list *t) { - struct sock *sk = (struct sock *)data; + struct inet_connection_sock *icsk = + from_timer(icsk, t, icsk_retransmit_timer); + struct sock *sk = &icsk->icsk_inet.sk; bh_lock_sock(sk); if (!sock_owned_by_user(sk)) { @@ -613,9 +617,9 @@ void tcp_set_keepalive(struct sock *sk, int val) EXPORT_SYMBOL_GPL(tcp_set_keepalive); -static void tcp_keepalive_timer (unsigned long data) +static void tcp_keepalive_timer (struct timer_list *t) { - struct sock *sk = (struct sock *) data; + struct sock *sk = from_timer(sk, t, sk_timer); struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); u32 elapsed; -- cgit v1.2.3 From 78802011fbe34331bdef6f2dfb1634011f0e4c32 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 16 Oct 2017 17:29:20 -0700 Subject: inet: frags: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: Alexander Aring Cc: Stefan Schmidt Cc: "David S. Miller" Cc: Alexey Kuznetsov Cc: Hideaki YOSHIFUJI Cc: Pablo Neira Ayuso Cc: Jozsef Kadlecsik Cc: Florian Westphal Cc: linux-wpan@vger.kernel.org Cc: netdev@vger.kernel.org Cc: netfilter-devel@vger.kernel.org Cc: coreteam@netfilter.org Signed-off-by: Kees Cook Acked-by: Stefan Schmidt # for ieee802154 Signed-off-by: David S. Miller --- include/net/inet_frag.h | 2 +- net/ieee802154/6lowpan/reassembly.c | 5 +++-- net/ipv4/inet_fragment.c | 4 ++-- net/ipv4/ip_fragment.c | 5 +++-- net/ipv6/netfilter/nf_conntrack_reasm.c | 5 +++-- net/ipv6/reassembly.c | 5 +++-- 6 files changed, 15 insertions(+), 11 deletions(-) (limited to 'include/net') diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h index fc59e0775e00..c695807ca707 100644 --- a/include/net/inet_frag.h +++ b/include/net/inet_frag.h @@ -95,7 +95,7 @@ struct inet_frags { void (*constructor)(struct inet_frag_queue *q, const void *arg); void (*destructor)(struct inet_frag_queue *); - void (*frag_expire)(unsigned long data); + void (*frag_expire)(struct timer_list *t); struct kmem_cache *frags_cachep; const char *frags_cache_name; }; diff --git a/net/ieee802154/6lowpan/reassembly.c b/net/ieee802154/6lowpan/reassembly.c index f85b08baff16..85bf86ad6b18 100644 --- a/net/ieee802154/6lowpan/reassembly.c +++ b/net/ieee802154/6lowpan/reassembly.c @@ -80,12 +80,13 @@ static void lowpan_frag_init(struct inet_frag_queue *q, const void *a) fq->daddr = *arg->dst; } -static void lowpan_frag_expire(unsigned long data) +static void lowpan_frag_expire(struct timer_list *t) { + struct inet_frag_queue *frag = from_timer(frag, t, timer); struct frag_queue *fq; struct net *net; - fq = container_of((struct inet_frag_queue *)data, struct frag_queue, q); + fq = container_of(frag, struct frag_queue, q); net = container_of(fq->q.net, struct net, ieee802154_lowpan.frags); spin_lock(&fq->q.lock); diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c index af74d0433453..7f3ef5c287a1 100644 --- a/net/ipv4/inet_fragment.c +++ b/net/ipv4/inet_fragment.c @@ -147,7 +147,7 @@ inet_evict_bucket(struct inet_frags *f, struct inet_frag_bucket *hb) spin_unlock(&hb->chain_lock); hlist_for_each_entry_safe(fq, n, &expired, list_evictor) - f->frag_expire((unsigned long) fq); + f->frag_expire(&fq->timer); return evicted; } @@ -366,7 +366,7 @@ static struct inet_frag_queue *inet_frag_alloc(struct netns_frags *nf, f->constructor(q, arg); add_frag_mem_limit(nf, f->qsize); - setup_timer(&q->timer, f->frag_expire, (unsigned long)q); + timer_setup(&q->timer, f->frag_expire, 0); spin_lock_init(&q->lock); refcount_set(&q->refcnt, 1); diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index 46408c220d9d..9215654a401f 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -190,12 +190,13 @@ static bool frag_expire_skip_icmp(u32 user) /* * Oops, a fragment queue timed out. Kill it and send an ICMP reply. */ -static void ip_expire(unsigned long arg) +static void ip_expire(struct timer_list *t) { + struct inet_frag_queue *frag = from_timer(frag, t, timer); struct ipq *qp; struct net *net; - qp = container_of((struct inet_frag_queue *) arg, struct ipq, q); + qp = container_of(frag, struct ipq, q); net = container_of(qp->q.net, struct net, ipv4.frags); rcu_read_lock(); diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index b263bf3a19f7..977d8900cfd1 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -169,12 +169,13 @@ static unsigned int nf_hashfn(const struct inet_frag_queue *q) return nf_hash_frag(nq->id, &nq->saddr, &nq->daddr); } -static void nf_ct_frag6_expire(unsigned long data) +static void nf_ct_frag6_expire(struct timer_list *t) { + struct inet_frag_queue *frag = from_timer(frag, t, timer); struct frag_queue *fq; struct net *net; - fq = container_of((struct inet_frag_queue *)data, struct frag_queue, q); + fq = container_of(frag, struct frag_queue, q); net = container_of(fq->q.net, struct net, nf_frag.frags); ip6_expire_frag_queue(net, fq, &nf_frags); diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index 846012eae526..afbc000ad4f2 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -170,12 +170,13 @@ out: } EXPORT_SYMBOL(ip6_expire_frag_queue); -static void ip6_frag_expire(unsigned long data) +static void ip6_frag_expire(struct timer_list *t) { + struct inet_frag_queue *frag = from_timer(frag, t, timer); struct frag_queue *fq; struct net *net; - fq = container_of((struct inet_frag_queue *)data, struct frag_queue, q); + fq = container_of(frag, struct frag_queue, q); net = container_of(fq->q.net, struct net, ipv6.frags); ip6_expire_frag_queue(net, fq, &ip6_frags); -- cgit v1.2.3 From de95e04791a03de5cb681980a3880db6919e3b4a Mon Sep 17 00:00:00 2001 From: David Ahern Date: Wed, 18 Oct 2017 09:56:54 -0700 Subject: net: Add extack to validator_info structs used for address notifier Add extack to in_validator_info and in6_validator_info. Update the one user of each, ipvlan, to return an error message for failures. Only manual configuration of an address is plumbed in the IPv6 code path. Signed-off-by: David Ahern Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller --- drivers/net/ipvlan/ipvlan_main.c | 10 ++++++++-- include/linux/inetdevice.h | 1 + include/net/addrconf.h | 1 + net/ipv4/devinet.c | 8 +++++--- net/ipv6/addrconf.c | 22 ++++++++++++---------- 5 files changed, 27 insertions(+), 15 deletions(-) (limited to 'include/net') diff --git a/drivers/net/ipvlan/ipvlan_main.c b/drivers/net/ipvlan/ipvlan_main.c index 6842739b6679..f0ab55df57f1 100644 --- a/drivers/net/ipvlan/ipvlan_main.c +++ b/drivers/net/ipvlan/ipvlan_main.c @@ -847,8 +847,11 @@ static int ipvlan_addr6_validator_event(struct notifier_block *unused, switch (event) { case NETDEV_UP: - if (ipvlan_addr_busy(ipvlan->port, &i6vi->i6vi_addr, true)) + if (ipvlan_addr_busy(ipvlan->port, &i6vi->i6vi_addr, true)) { + NL_SET_ERR_MSG(i6vi->extack, + "Address already assigned to an ipvlan device"); return notifier_from_errno(-EADDRINUSE); + } break; } @@ -917,8 +920,11 @@ static int ipvlan_addr4_validator_event(struct notifier_block *unused, switch (event) { case NETDEV_UP: - if (ipvlan_addr_busy(ipvlan->port, &ivi->ivi_addr, false)) + if (ipvlan_addr_busy(ipvlan->port, &ivi->ivi_addr, false)) { + NL_SET_ERR_MSG(ivi->extack, + "Address already assigned to an ipvlan device"); return notifier_from_errno(-EADDRINUSE); + } break; } diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h index 751d051f0bc7..681dff30940b 100644 --- a/include/linux/inetdevice.h +++ b/include/linux/inetdevice.h @@ -154,6 +154,7 @@ struct in_ifaddr { struct in_validator_info { __be32 ivi_addr; struct in_device *ivi_dev; + struct netlink_ext_ack *extack; }; int register_inetaddr_notifier(struct notifier_block *nb); diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 87981cd63180..b8b16437c6d5 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -55,6 +55,7 @@ struct prefix_info { struct in6_validator_info { struct in6_addr i6vi_addr; struct inet6_dev *i6vi_dev; + struct netlink_ext_ack *extack; }; #define IN6_ADDR_HSIZE_SHIFT 4 diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index e1e2ec0525e6..a4573bccd6da 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -444,7 +444,7 @@ static void check_lifetime(struct work_struct *work); static DECLARE_DELAYED_WORK(check_lifetime_work, check_lifetime); static int __inet_insert_ifa(struct in_ifaddr *ifa, struct nlmsghdr *nlh, - u32 portid) + u32 portid, struct netlink_ext_ack *extack) { struct in_device *in_dev = ifa->ifa_dev; struct in_ifaddr *ifa1, **ifap, **last_primary; @@ -489,6 +489,7 @@ static int __inet_insert_ifa(struct in_ifaddr *ifa, struct nlmsghdr *nlh, */ ivi.ivi_addr = ifa->ifa_address; ivi.ivi_dev = ifa->ifa_dev; + ivi.extack = extack; ret = blocking_notifier_call_chain(&inetaddr_validator_chain, NETDEV_UP, &ivi); ret = notifier_to_errno(ret); @@ -521,7 +522,7 @@ static int __inet_insert_ifa(struct in_ifaddr *ifa, struct nlmsghdr *nlh, static int inet_insert_ifa(struct in_ifaddr *ifa) { - return __inet_insert_ifa(ifa, NULL, 0); + return __inet_insert_ifa(ifa, NULL, 0, NULL); } static int inet_set_ifa(struct net_device *dev, struct in_ifaddr *ifa) @@ -902,7 +903,8 @@ static int inet_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, return ret; } } - return __inet_insert_ifa(ifa, nlh, NETLINK_CB(skb).portid); + return __inet_insert_ifa(ifa, nlh, NETLINK_CB(skb).portid, + extack); } else { inet_free_ifa(ifa); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index dd9c0c435f71..93f9c0a61911 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -987,7 +987,7 @@ static struct inet6_ifaddr * ipv6_add_addr(struct inet6_dev *idev, const struct in6_addr *addr, const struct in6_addr *peer_addr, int pfxlen, int scope, u32 flags, u32 valid_lft, u32 prefered_lft, - bool can_block) + bool can_block, struct netlink_ext_ack *extack) { gfp_t gfp_flags = can_block ? GFP_KERNEL : GFP_ATOMIC; struct net *net = dev_net(idev->dev); @@ -1019,6 +1019,7 @@ ipv6_add_addr(struct inet6_dev *idev, const struct in6_addr *addr, struct in6_validator_info i6vi = { .i6vi_addr = *addr, .i6vi_dev = idev, + .extack = extack, }; err = inet6addr_validator_notifier_call_chain(NETDEV_UP, &i6vi); @@ -1356,7 +1357,7 @@ retry: ift = ipv6_add_addr(idev, &addr, NULL, tmp_plen, ipv6_addr_scope(&addr), addr_flags, - tmp_valid_lft, tmp_prefered_lft, true); + tmp_valid_lft, tmp_prefered_lft, true, NULL); if (IS_ERR(ift)) { in6_ifa_put(ifp); in6_dev_put(idev); @@ -2040,7 +2041,7 @@ void addrconf_dad_failure(struct inet6_ifaddr *ifp) ifp2 = ipv6_add_addr(idev, &new_addr, NULL, pfxlen, scope, flags, valid_lft, - preferred_lft, false); + preferred_lft, false, NULL); if (IS_ERR(ifp2)) goto lock_errdad; @@ -2498,7 +2499,7 @@ int addrconf_prefix_rcv_add_addr(struct net *net, struct net_device *dev, pinfo->prefix_len, addr_type&IPV6_ADDR_SCOPE_MASK, addr_flags, valid_lft, - prefered_lft, false); + prefered_lft, false, NULL); if (IS_ERR_OR_NULL(ifp)) return -1; @@ -2808,7 +2809,8 @@ static int inet6_addr_add(struct net *net, int ifindex, const struct in6_addr *pfx, const struct in6_addr *peer_pfx, unsigned int plen, __u32 ifa_flags, - __u32 prefered_lft, __u32 valid_lft) + __u32 prefered_lft, __u32 valid_lft, + struct netlink_ext_ack *extack) { struct inet6_ifaddr *ifp; struct inet6_dev *idev; @@ -2867,7 +2869,7 @@ static int inet6_addr_add(struct net *net, int ifindex, } ifp = ipv6_add_addr(idev, pfx, peer_pfx, plen, scope, ifa_flags, - valid_lft, prefered_lft, true); + valid_lft, prefered_lft, true, extack); if (!IS_ERR(ifp)) { if (!(ifa_flags & IFA_F_NOPREFIXROUTE)) { @@ -2952,7 +2954,7 @@ int addrconf_add_ifaddr(struct net *net, void __user *arg) rtnl_lock(); err = inet6_addr_add(net, ireq.ifr6_ifindex, &ireq.ifr6_addr, NULL, ireq.ifr6_prefixlen, IFA_F_PERMANENT, - INFINITY_LIFE_TIME, INFINITY_LIFE_TIME); + INFINITY_LIFE_TIME, INFINITY_LIFE_TIME, NULL); rtnl_unlock(); return err; } @@ -2983,7 +2985,7 @@ static void add_addr(struct inet6_dev *idev, const struct in6_addr *addr, ifp = ipv6_add_addr(idev, addr, NULL, plen, scope, IFA_F_PERMANENT, INFINITY_LIFE_TIME, INFINITY_LIFE_TIME, - true); + true, NULL); if (!IS_ERR(ifp)) { spin_lock_bh(&ifp->lock); ifp->flags &= ~IFA_F_TENTATIVE; @@ -3083,7 +3085,7 @@ void addrconf_add_linklocal(struct inet6_dev *idev, #endif ifp = ipv6_add_addr(idev, addr, NULL, 64, IFA_LINK, addr_flags, - INFINITY_LIFE_TIME, INFINITY_LIFE_TIME, true); + INFINITY_LIFE_TIME, INFINITY_LIFE_TIME, true, NULL); if (!IS_ERR(ifp)) { addrconf_prefix_route(&ifp->addr, ifp->prefix_len, idev->dev, 0, 0); addrconf_dad_start(ifp); @@ -4586,7 +4588,7 @@ inet6_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, */ return inet6_addr_add(net, ifm->ifa_index, pfx, peer_pfx, ifm->ifa_prefixlen, ifa_flags, - preferred_lft, valid_lft); + preferred_lft, valid_lft, extack); } if (nlh->nlmsg_flags & NLM_F_EXCL || -- cgit v1.2.3 From 1fba70e5b6bed53496ba1f1f16127f5be01b5fb6 Mon Sep 17 00:00:00 2001 From: Yuchung Cheng Date: Wed, 18 Oct 2017 11:22:51 -0700 Subject: tcp: socket option to set TCP fast open key New socket option TCP_FASTOPEN_KEY to allow different keys per listener. The listener by default uses the global key until the socket option is set. The key is a 16 bytes long binary data. This option has no effect on regular non-listener TCP sockets. Signed-off-by: Yuchung Cheng Reviewed-by: Eric Dumazet Reviewed-by: Christoph Paasch Signed-off-by: David S. Miller --- include/net/request_sock.h | 2 ++ include/net/tcp.h | 5 +++-- include/uapi/linux/tcp.h | 1 + net/ipv4/sysctl_net_ipv4.c | 3 ++- net/ipv4/tcp.c | 33 +++++++++++++++++++++++++++ net/ipv4/tcp_fastopen.c | 56 +++++++++++++++++++++++++++++++++------------- net/ipv4/tcp_ipv4.c | 1 + 7 files changed, 82 insertions(+), 19 deletions(-) (limited to 'include/net') diff --git a/include/net/request_sock.h b/include/net/request_sock.h index 23e22054aa60..347015515a7d 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -150,6 +150,8 @@ struct fastopen_queue { spinlock_t lock; int qlen; /* # of pending (TCP_SYN_RECV) reqs */ int max_qlen; /* != 0 iff TFO is currently enabled */ + + struct tcp_fastopen_context __rcu *ctx; /* cipher context for cookie */ }; /** struct request_sock_queue - queue of request_socks diff --git a/include/net/tcp.h b/include/net/tcp.h index 3b3b9b968e2d..1efe8365cb28 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1555,9 +1555,10 @@ struct tcp_fastopen_request { int copied; /* queued in tcp_connect() */ }; void tcp_free_fastopen_req(struct tcp_sock *tp); - +void tcp_fastopen_destroy_cipher(struct sock *sk); void tcp_fastopen_ctx_destroy(struct net *net); -int tcp_fastopen_reset_cipher(struct net *net, void *key, unsigned int len); +int tcp_fastopen_reset_cipher(struct net *net, struct sock *sk, + void *key, unsigned int len); void tcp_fastopen_add_skb(struct sock *sk, struct sk_buff *skb); struct sock *tcp_try_fastopen(struct sock *sk, struct sk_buff *skb, struct request_sock *req, diff --git a/include/uapi/linux/tcp.h b/include/uapi/linux/tcp.h index 15c25eccab2b..69c7493e42f8 100644 --- a/include/uapi/linux/tcp.h +++ b/include/uapi/linux/tcp.h @@ -119,6 +119,7 @@ enum { #define TCP_FASTOPEN_CONNECT 30 /* Attempt FastOpen with connect */ #define TCP_ULP 31 /* Attach a ULP to a TCP connection */ #define TCP_MD5SIG_EXT 32 /* TCP MD5 Signature with extensions */ +#define TCP_FASTOPEN_KEY 33 /* Set the key for Fast Open (cookie) */ struct tcp_repair_opt { __u32 opt_code; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index cac8dd309f39..81d218346cf7 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -284,7 +284,8 @@ static int proc_tcp_fastopen_key(struct ctl_table *table, int write, ret = -EINVAL; goto bad_key; } - tcp_fastopen_reset_cipher(net, user_key, TCP_FASTOPEN_KEY_LENGTH); + tcp_fastopen_reset_cipher(net, NULL, user_key, + TCP_FASTOPEN_KEY_LENGTH); } bad_key: diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 3b34850d361f..8b1fa4dd4538 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2571,6 +2571,17 @@ static int do_tcp_setsockopt(struct sock *sk, int level, release_sock(sk); return err; } + case TCP_FASTOPEN_KEY: { + __u8 key[TCP_FASTOPEN_KEY_LENGTH]; + + if (optlen != sizeof(key)) + return -EINVAL; + + if (copy_from_user(key, optval, optlen)) + return -EFAULT; + + return tcp_fastopen_reset_cipher(net, sk, key, sizeof(key)); + } default: /* fallthru */ break; @@ -3157,6 +3168,28 @@ static int do_tcp_getsockopt(struct sock *sk, int level, return -EFAULT; return 0; + case TCP_FASTOPEN_KEY: { + __u8 key[TCP_FASTOPEN_KEY_LENGTH]; + struct tcp_fastopen_context *ctx; + + if (get_user(len, optlen)) + return -EFAULT; + + rcu_read_lock(); + ctx = rcu_dereference(icsk->icsk_accept_queue.fastopenq.ctx); + if (ctx) + memcpy(key, ctx->key, sizeof(key)); + else + len = 0; + rcu_read_unlock(); + + len = min_t(unsigned int, len, sizeof(key)); + if (put_user(len, optlen)) + return -EFAULT; + if (copy_to_user(optval, key, len)) + return -EFAULT; + return 0; + } case TCP_THIN_LINEAR_TIMEOUTS: val = tp->thin_lto; break; diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 7ee4aadcdd71..21075ce19cb6 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -29,7 +29,7 @@ void tcp_fastopen_init_key_once(struct net *net) * for a valid cookie, so this is an acceptable risk. */ get_random_bytes(key, sizeof(key)); - tcp_fastopen_reset_cipher(net, key, sizeof(key)); + tcp_fastopen_reset_cipher(net, NULL, key, sizeof(key)); } static void tcp_fastopen_ctx_free(struct rcu_head *head) @@ -40,6 +40,16 @@ static void tcp_fastopen_ctx_free(struct rcu_head *head) kfree(ctx); } +void tcp_fastopen_destroy_cipher(struct sock *sk) +{ + struct tcp_fastopen_context *ctx; + + ctx = rcu_dereference_protected( + inet_csk(sk)->icsk_accept_queue.fastopenq.ctx, 1); + if (ctx) + call_rcu(&ctx->rcu, tcp_fastopen_ctx_free); +} + void tcp_fastopen_ctx_destroy(struct net *net) { struct tcp_fastopen_context *ctxt; @@ -55,10 +65,12 @@ void tcp_fastopen_ctx_destroy(struct net *net) call_rcu(&ctxt->rcu, tcp_fastopen_ctx_free); } -int tcp_fastopen_reset_cipher(struct net *net, void *key, unsigned int len) +int tcp_fastopen_reset_cipher(struct net *net, struct sock *sk, + void *key, unsigned int len) { - int err; struct tcp_fastopen_context *ctx, *octx; + struct fastopen_queue *q; + int err; ctx = kmalloc(sizeof(*ctx), GFP_KERNEL); if (!ctx) @@ -79,27 +91,39 @@ error: kfree(ctx); } memcpy(ctx->key, key, len); - spin_lock(&net->ipv4.tcp_fastopen_ctx_lock); - octx = rcu_dereference_protected(net->ipv4.tcp_fastopen_ctx, - lockdep_is_held(&net->ipv4.tcp_fastopen_ctx_lock)); - rcu_assign_pointer(net->ipv4.tcp_fastopen_ctx, ctx); - spin_unlock(&net->ipv4.tcp_fastopen_ctx_lock); + if (sk) { + q = &inet_csk(sk)->icsk_accept_queue.fastopenq; + spin_lock_bh(&q->lock); + octx = rcu_dereference_protected(q->ctx, + lockdep_is_held(&q->lock)); + rcu_assign_pointer(q->ctx, ctx); + spin_unlock_bh(&q->lock); + } else { + spin_lock(&net->ipv4.tcp_fastopen_ctx_lock); + octx = rcu_dereference_protected(net->ipv4.tcp_fastopen_ctx, + lockdep_is_held(&net->ipv4.tcp_fastopen_ctx_lock)); + rcu_assign_pointer(net->ipv4.tcp_fastopen_ctx, ctx); + spin_unlock(&net->ipv4.tcp_fastopen_ctx_lock); + } if (octx) call_rcu(&octx->rcu, tcp_fastopen_ctx_free); return err; } -static bool __tcp_fastopen_cookie_gen(struct net *net, - const void *path, +static bool __tcp_fastopen_cookie_gen(struct sock *sk, const void *path, struct tcp_fastopen_cookie *foc) { struct tcp_fastopen_context *ctx; bool ok = false; rcu_read_lock(); - ctx = rcu_dereference(net->ipv4.tcp_fastopen_ctx); + + ctx = rcu_dereference(inet_csk(sk)->icsk_accept_queue.fastopenq.ctx); + if (!ctx) + ctx = rcu_dereference(sock_net(sk)->ipv4.tcp_fastopen_ctx); + if (ctx) { crypto_cipher_encrypt_one(ctx->tfm, foc->val, path); foc->len = TCP_FASTOPEN_COOKIE_SIZE; @@ -115,7 +139,7 @@ static bool __tcp_fastopen_cookie_gen(struct net *net, * * XXX (TFO) - refactor when TCP_FASTOPEN_COOKIE_SIZE != AES_BLOCK_SIZE. */ -static bool tcp_fastopen_cookie_gen(struct net *net, +static bool tcp_fastopen_cookie_gen(struct sock *sk, struct request_sock *req, struct sk_buff *syn, struct tcp_fastopen_cookie *foc) @@ -124,7 +148,7 @@ static bool tcp_fastopen_cookie_gen(struct net *net, const struct iphdr *iph = ip_hdr(syn); __be32 path[4] = { iph->saddr, iph->daddr, 0, 0 }; - return __tcp_fastopen_cookie_gen(net, path, foc); + return __tcp_fastopen_cookie_gen(sk, path, foc); } #if IS_ENABLED(CONFIG_IPV6) @@ -132,13 +156,13 @@ static bool tcp_fastopen_cookie_gen(struct net *net, const struct ipv6hdr *ip6h = ipv6_hdr(syn); struct tcp_fastopen_cookie tmp; - if (__tcp_fastopen_cookie_gen(net, &ip6h->saddr, &tmp)) { + if (__tcp_fastopen_cookie_gen(sk, &ip6h->saddr, &tmp)) { struct in6_addr *buf = &tmp.addr; int i; for (i = 0; i < 4; i++) buf->s6_addr32[i] ^= ip6h->daddr.s6_addr32[i]; - return __tcp_fastopen_cookie_gen(net, buf, foc); + return __tcp_fastopen_cookie_gen(sk, buf, foc); } } #endif @@ -313,7 +337,7 @@ struct sock *tcp_try_fastopen(struct sock *sk, struct sk_buff *skb, goto fastopen; if (foc->len >= 0 && /* Client presents or requests a cookie */ - tcp_fastopen_cookie_gen(sock_net(sk), req, skb, &valid_foc) && + tcp_fastopen_cookie_gen(sk, req, skb, &valid_foc) && foc->len == TCP_FASTOPEN_COOKIE_SIZE && foc->len == valid_foc.len && !memcmp(foc->val, valid_foc.val, foc->len)) { diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index ecee4ddb24c5..28ca4e177047 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1893,6 +1893,7 @@ void tcp_v4_destroy_sock(struct sock *sk) /* If socket is aborted during connect operation */ tcp_free_fastopen_req(tp); + tcp_fastopen_destroy_cipher(sk); tcp_saved_syn_free(tp); sk_sockets_allocated_dec(sk); -- cgit v1.2.3 From 8c4083b30e56fc71b0e94c26374b32d95d5ea461 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 19 Oct 2017 15:50:29 +0200 Subject: net: sched: add block bind/unbind notif. and extended block_get/put Introduce new type of ndo_setup_tc message to propage binding/unbinding of a block to driver. Call this ndo whenever qdisc gets/puts a block. Alongside with this, there's need to propagate binder type from qdisc code down to the notifier. So introduce extended variants of block_get/put in order to pass this info. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/linux/netdevice.h | 1 + include/net/pkt_cls.h | 40 +++++++++++++++++++++++++++++++++ net/sched/cls_api.c | 56 ++++++++++++++++++++++++++++++++++++++++++++--- 3 files changed, 94 insertions(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index bf014afcb914..4de5b08ee0fb 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -775,6 +775,7 @@ enum tc_setup_type { TC_SETUP_CLSFLOWER, TC_SETUP_CLSMATCHALL, TC_SETUP_CLSBPF, + TC_SETUP_BLOCK, }; /* These structures hold the attributes of xdp state that are being passed diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index 49a143e0fe65..41bc7d774047 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -17,13 +17,27 @@ struct tcf_walker { int register_tcf_proto_ops(struct tcf_proto_ops *ops); int unregister_tcf_proto_ops(struct tcf_proto_ops *ops); +enum tcf_block_binder_type { + TCF_BLOCK_BINDER_TYPE_UNSPEC, +}; + +struct tcf_block_ext_info { + enum tcf_block_binder_type binder_type; +}; + #ifdef CONFIG_NET_CLS struct tcf_chain *tcf_chain_get(struct tcf_block *block, u32 chain_index, bool create); void tcf_chain_put(struct tcf_chain *chain); int tcf_block_get(struct tcf_block **p_block, struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q); +int tcf_block_get_ext(struct tcf_block **p_block, + struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q, + struct tcf_block_ext_info *ei); void tcf_block_put(struct tcf_block *block); +void tcf_block_put_ext(struct tcf_block *block, + struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q, + struct tcf_block_ext_info *ei); static inline struct Qdisc *tcf_block_q(struct tcf_block *block) { @@ -46,10 +60,25 @@ int tcf_block_get(struct tcf_block **p_block, return 0; } +static inline +int tcf_block_get_ext(struct tcf_block **p_block, + struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q, + struct tcf_block_ext_info *ei) +{ + return 0; +} + static inline void tcf_block_put(struct tcf_block *block) { } +static inline +void tcf_block_put_ext(struct tcf_block *block, + struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q, + struct tcf_block_ext_info *ei) +{ +} + static inline struct Qdisc *tcf_block_q(struct tcf_block *block) { return NULL; @@ -434,6 +463,17 @@ tcf_match_indev(struct sk_buff *skb, int ifindex) int tc_setup_cb_call(struct tcf_exts *exts, enum tc_setup_type type, void *type_data, bool err_stop); +enum tc_block_command { + TC_BLOCK_BIND, + TC_BLOCK_UNBIND, +}; + +struct tc_block_offload { + enum tc_block_command command; + enum tcf_block_binder_type binder_type; + struct tcf_block *block; +}; + struct tc_cls_common_offload { u32 chain_index; __be16 protocol; diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 2e8e87fd9d97..92dce26d10e3 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -240,8 +240,36 @@ tcf_chain_filter_chain_ptr_set(struct tcf_chain *chain, chain->p_filter_chain = p_filter_chain; } -int tcf_block_get(struct tcf_block **p_block, - struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q) +static void tcf_block_offload_cmd(struct tcf_block *block, struct Qdisc *q, + struct tcf_block_ext_info *ei, + enum tc_block_command command) +{ + struct net_device *dev = q->dev_queue->dev; + struct tc_block_offload bo = {}; + + if (!tc_can_offload(dev)) + return; + bo.command = command; + bo.binder_type = ei->binder_type; + bo.block = block; + dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_BLOCK, &bo); +} + +static void tcf_block_offload_bind(struct tcf_block *block, struct Qdisc *q, + struct tcf_block_ext_info *ei) +{ + tcf_block_offload_cmd(block, q, ei, TC_BLOCK_BIND); +} + +static void tcf_block_offload_unbind(struct tcf_block *block, struct Qdisc *q, + struct tcf_block_ext_info *ei) +{ + tcf_block_offload_cmd(block, q, ei, TC_BLOCK_UNBIND); +} + +int tcf_block_get_ext(struct tcf_block **p_block, + struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q, + struct tcf_block_ext_info *ei) { struct tcf_block *block = kzalloc(sizeof(*block), GFP_KERNEL); struct tcf_chain *chain; @@ -259,6 +287,7 @@ int tcf_block_get(struct tcf_block **p_block, tcf_chain_filter_chain_ptr_set(chain, p_filter_chain); block->net = qdisc_net(q); block->q = q; + tcf_block_offload_bind(block, q, ei); *p_block = block; return 0; @@ -266,15 +295,28 @@ err_chain_create: kfree(block); return err; } +EXPORT_SYMBOL(tcf_block_get_ext); + +int tcf_block_get(struct tcf_block **p_block, + struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q) +{ + struct tcf_block_ext_info ei = {0, }; + + return tcf_block_get_ext(p_block, p_filter_chain, q, &ei); +} EXPORT_SYMBOL(tcf_block_get); -void tcf_block_put(struct tcf_block *block) +void tcf_block_put_ext(struct tcf_block *block, + struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q, + struct tcf_block_ext_info *ei) { struct tcf_chain *chain, *tmp; if (!block) return; + tcf_block_offload_unbind(block, q, ei); + /* XXX: Standalone actions are not allowed to jump to any chain, and * bound actions should be all removed after flushing. However, * filters are destroyed in RCU callbacks, we have to hold the chains @@ -302,6 +344,14 @@ void tcf_block_put(struct tcf_block *block) tcf_chain_put(chain); kfree(block); } +EXPORT_SYMBOL(tcf_block_put_ext); + +void tcf_block_put(struct tcf_block *block) +{ + struct tcf_block_ext_info ei = {0, }; + + tcf_block_put_ext(block, NULL, block->q, &ei); +} EXPORT_SYMBOL(tcf_block_put); /* Main classifier routine: scans classifier chain attached -- cgit v1.2.3 From 6e40cf2d4dee9dc22ff398041ce876bef8172dea Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 19 Oct 2017 15:50:30 +0200 Subject: net: sched: use extended variants of block_get/put in ingress and clsact qdiscs Use previously introduced extended variants of block get and put functions. This allows to specify a binder types specific to clsact ingress/egress which is useful for drivers to distinguish who actually got the block. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/pkt_cls.h | 2 ++ net/sched/sch_ingress.c | 36 +++++++++++++++++++++++++++++------- 2 files changed, 31 insertions(+), 7 deletions(-) (limited to 'include/net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index 41bc7d774047..5c50af8f7183 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -19,6 +19,8 @@ int unregister_tcf_proto_ops(struct tcf_proto_ops *ops); enum tcf_block_binder_type { TCF_BLOCK_BINDER_TYPE_UNSPEC, + TCF_BLOCK_BINDER_TYPE_CLSACT_INGRESS, + TCF_BLOCK_BINDER_TYPE_CLSACT_EGRESS, }; struct tcf_block_ext_info { diff --git a/net/sched/sch_ingress.c b/net/sched/sch_ingress.c index 9ccc1b89b0d9..b599db26d34b 100644 --- a/net/sched/sch_ingress.c +++ b/net/sched/sch_ingress.c @@ -20,6 +20,7 @@ struct ingress_sched_data { struct tcf_block *block; + struct tcf_block_ext_info block_info; }; static struct Qdisc *ingress_leaf(struct Qdisc *sch, unsigned long arg) @@ -59,7 +60,10 @@ static int ingress_init(struct Qdisc *sch, struct nlattr *opt) struct net_device *dev = qdisc_dev(sch); int err; - err = tcf_block_get(&q->block, &dev->ingress_cl_list, sch); + q->block_info.binder_type = TCF_BLOCK_BINDER_TYPE_CLSACT_INGRESS; + + err = tcf_block_get_ext(&q->block, &dev->ingress_cl_list, + sch, &q->block_info); if (err) return err; @@ -72,8 +76,10 @@ static int ingress_init(struct Qdisc *sch, struct nlattr *opt) static void ingress_destroy(struct Qdisc *sch) { struct ingress_sched_data *q = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); - tcf_block_put(q->block); + tcf_block_put_ext(q->block, &dev->ingress_cl_list, + sch, &q->block_info); net_dec_ingress_queue(); } @@ -114,6 +120,8 @@ static struct Qdisc_ops ingress_qdisc_ops __read_mostly = { struct clsact_sched_data { struct tcf_block *ingress_block; struct tcf_block *egress_block; + struct tcf_block_ext_info ingress_block_info; + struct tcf_block_ext_info egress_block_info; }; static unsigned long clsact_find(struct Qdisc *sch, u32 classid) @@ -153,13 +161,19 @@ static int clsact_init(struct Qdisc *sch, struct nlattr *opt) struct net_device *dev = qdisc_dev(sch); int err; - err = tcf_block_get(&q->ingress_block, &dev->ingress_cl_list, sch); + q->ingress_block_info.binder_type = TCF_BLOCK_BINDER_TYPE_CLSACT_INGRESS; + + err = tcf_block_get_ext(&q->ingress_block, &dev->ingress_cl_list, + sch, &q->ingress_block_info); if (err) return err; - err = tcf_block_get(&q->egress_block, &dev->egress_cl_list, sch); + q->egress_block_info.binder_type = TCF_BLOCK_BINDER_TYPE_CLSACT_EGRESS; + + err = tcf_block_get_ext(&q->egress_block, &dev->egress_cl_list, + sch, &q->egress_block_info); if (err) - return err; + goto err_egress_block_get; net_inc_ingress_queue(); net_inc_egress_queue(); @@ -167,14 +181,22 @@ static int clsact_init(struct Qdisc *sch, struct nlattr *opt) sch->flags |= TCQ_F_CPUSTATS; return 0; + +err_egress_block_get: + tcf_block_put_ext(q->ingress_block, &dev->ingress_cl_list, + sch, &q->ingress_block_info); + return err; } static void clsact_destroy(struct Qdisc *sch) { struct clsact_sched_data *q = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); - tcf_block_put(q->egress_block); - tcf_block_put(q->ingress_block); + tcf_block_put_ext(q->egress_block, &dev->egress_cl_list, + sch, &q->egress_block_info); + tcf_block_put_ext(q->ingress_block, &dev->ingress_cl_list, + sch, &q->ingress_block_info); net_dec_ingress_queue(); net_dec_egress_queue(); -- cgit v1.2.3 From acb674428c3d57bccbe3f4a1a7a009f6d73e9f41 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 19 Oct 2017 15:50:31 +0200 Subject: net: sched: introduce per-block callbacks Introduce infrastructure that allows drivers to register callbacks that are called whenever tc would offload inserted rule for a specific block. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/pkt_cls.h | 81 +++++++++++++++++++++++++++++++++++ include/net/sch_generic.h | 1 + net/sched/cls_api.c | 105 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 187 insertions(+) (limited to 'include/net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index 5c50af8f7183..4bc6b1cc245d 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -27,6 +27,8 @@ struct tcf_block_ext_info { enum tcf_block_binder_type binder_type; }; +struct tcf_block_cb; + #ifdef CONFIG_NET_CLS struct tcf_chain *tcf_chain_get(struct tcf_block *block, u32 chain_index, bool create); @@ -51,6 +53,21 @@ static inline struct net_device *tcf_block_dev(struct tcf_block *block) return tcf_block_q(block)->dev_queue->dev; } +void *tcf_block_cb_priv(struct tcf_block_cb *block_cb); +struct tcf_block_cb *tcf_block_cb_lookup(struct tcf_block *block, + tc_setup_cb_t *cb, void *cb_ident); +void tcf_block_cb_incref(struct tcf_block_cb *block_cb); +unsigned int tcf_block_cb_decref(struct tcf_block_cb *block_cb); +struct tcf_block_cb *__tcf_block_cb_register(struct tcf_block *block, + tc_setup_cb_t *cb, void *cb_ident, + void *cb_priv); +int tcf_block_cb_register(struct tcf_block *block, + tc_setup_cb_t *cb, void *cb_ident, + void *cb_priv); +void __tcf_block_cb_unregister(struct tcf_block_cb *block_cb); +void tcf_block_cb_unregister(struct tcf_block *block, + tc_setup_cb_t *cb, void *cb_ident); + int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp, struct tcf_result *res, bool compat_mode); @@ -91,6 +108,70 @@ static inline struct net_device *tcf_block_dev(struct tcf_block *block) return NULL; } +static inline +int tc_setup_cb_block_register(struct tcf_block *block, tc_setup_cb_t *cb, + void *cb_priv) +{ + return 0; +} + +static inline +void tc_setup_cb_block_unregister(struct tcf_block *block, tc_setup_cb_t *cb, + void *cb_priv) +{ +} + +static inline +void *tcf_block_cb_priv(struct tcf_block_cb *block_cb) +{ + return NULL; +} + +static inline +struct tcf_block_cb *tcf_block_cb_lookup(struct tcf_block *block, + tc_setup_cb_t *cb, void *cb_ident) +{ + return NULL; +} + +static inline +void tcf_block_cb_incref(struct tcf_block_cb *block_cb) +{ +} + +static inline +unsigned int tcf_block_cb_decref(struct tcf_block_cb *block_cb) +{ + return 0; +} + +static inline +struct tcf_block_cb *__tcf_block_cb_register(struct tcf_block *block, + tc_setup_cb_t *cb, void *cb_ident, + void *cb_priv) +{ + return NULL; +} + +static inline +int tcf_block_cb_register(struct tcf_block *block, + tc_setup_cb_t *cb, void *cb_ident, + void *cb_priv) +{ + return 0; +} + +static inline +void __tcf_block_cb_unregister(struct tcf_block_cb *block_cb) +{ +} + +static inline +void tcf_block_cb_unregister(struct tcf_block *block, + tc_setup_cb_t *cb, void *cb_ident) +{ +} + static inline int tcf_classify(struct sk_buff *skb, const struct tcf_proto *tp, struct tcf_result *res, bool compat_mode) { diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 0aea9e23e97a..031dffd5836c 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -272,6 +272,7 @@ struct tcf_block { struct list_head chain_list; struct net *net; struct Qdisc *q; + struct list_head cb_list; }; static inline void qdisc_cb_private_validate(const struct sk_buff *skb, int sz) diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 92dce26d10e3..b16c79c47be5 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -278,6 +278,8 @@ int tcf_block_get_ext(struct tcf_block **p_block, if (!block) return -ENOMEM; INIT_LIST_HEAD(&block->chain_list); + INIT_LIST_HEAD(&block->cb_list); + /* Create chain 0 by default, it has to be always present. */ chain = tcf_chain_create(block, 0); if (!chain) { @@ -354,6 +356,109 @@ void tcf_block_put(struct tcf_block *block) } EXPORT_SYMBOL(tcf_block_put); +struct tcf_block_cb { + struct list_head list; + tc_setup_cb_t *cb; + void *cb_ident; + void *cb_priv; + unsigned int refcnt; +}; + +void *tcf_block_cb_priv(struct tcf_block_cb *block_cb) +{ + return block_cb->cb_priv; +} +EXPORT_SYMBOL(tcf_block_cb_priv); + +struct tcf_block_cb *tcf_block_cb_lookup(struct tcf_block *block, + tc_setup_cb_t *cb, void *cb_ident) +{ struct tcf_block_cb *block_cb; + + list_for_each_entry(block_cb, &block->cb_list, list) + if (block_cb->cb == cb && block_cb->cb_ident == cb_ident) + return block_cb; + return NULL; +} +EXPORT_SYMBOL(tcf_block_cb_lookup); + +void tcf_block_cb_incref(struct tcf_block_cb *block_cb) +{ + block_cb->refcnt++; +} +EXPORT_SYMBOL(tcf_block_cb_incref); + +unsigned int tcf_block_cb_decref(struct tcf_block_cb *block_cb) +{ + return --block_cb->refcnt; +} +EXPORT_SYMBOL(tcf_block_cb_decref); + +struct tcf_block_cb *__tcf_block_cb_register(struct tcf_block *block, + tc_setup_cb_t *cb, void *cb_ident, + void *cb_priv) +{ + struct tcf_block_cb *block_cb; + + block_cb = kzalloc(sizeof(*block_cb), GFP_KERNEL); + if (!block_cb) + return NULL; + block_cb->cb = cb; + block_cb->cb_ident = cb_ident; + block_cb->cb_priv = cb_priv; + list_add(&block_cb->list, &block->cb_list); + return block_cb; +} +EXPORT_SYMBOL(__tcf_block_cb_register); + +int tcf_block_cb_register(struct tcf_block *block, + tc_setup_cb_t *cb, void *cb_ident, + void *cb_priv) +{ + struct tcf_block_cb *block_cb; + + block_cb = __tcf_block_cb_register(block, cb, cb_ident, cb_priv); + return block_cb ? 0 : -ENOMEM; +} +EXPORT_SYMBOL(tcf_block_cb_register); + +void __tcf_block_cb_unregister(struct tcf_block_cb *block_cb) +{ + list_del(&block_cb->list); + kfree(block_cb); +} +EXPORT_SYMBOL(__tcf_block_cb_unregister); + +void tcf_block_cb_unregister(struct tcf_block *block, + tc_setup_cb_t *cb, void *cb_ident) +{ + struct tcf_block_cb *block_cb; + + block_cb = tcf_block_cb_lookup(block, cb, cb_ident); + if (!block_cb) + return; + __tcf_block_cb_unregister(block_cb); +} +EXPORT_SYMBOL(tcf_block_cb_unregister); + +static int tcf_block_cb_call(struct tcf_block *block, enum tc_setup_type type, + void *type_data, bool err_stop) +{ + struct tcf_block_cb *block_cb; + int ok_count = 0; + int err; + + list_for_each_entry(block_cb, &block->cb_list, list) { + err = block_cb->cb(type, type_data, block_cb->cb_priv); + if (err) { + if (err_stop) + return err; + } else { + ok_count++; + } + } + return ok_count; +} + /* Main classifier routine: scans classifier chain attached * to this qdisc, (optionally) tests for protocol and asks * specific classifiers. -- cgit v1.2.3 From 208c0f4b5237f1d6611b2c679a8022d6901577d6 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 19 Oct 2017 15:50:32 +0200 Subject: net: sched: use tc_setup_cb_call to call per-block callbacks Extend the tc_setup_cb_call entrypoint function originally used only for action egress devices callbacks to call per-block callbacks as well. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/pkt_cls.h | 4 ++-- net/sched/cls_api.c | 21 ++++++++++++++++++--- net/sched/cls_flower.c | 9 ++++++--- 3 files changed, 26 insertions(+), 8 deletions(-) (limited to 'include/net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index 4bc6b1cc245d..fcca5a9d9880 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -543,8 +543,8 @@ tcf_match_indev(struct sk_buff *skb, int ifindex) } #endif /* CONFIG_NET_CLS_IND */ -int tc_setup_cb_call(struct tcf_exts *exts, enum tc_setup_type type, - void *type_data, bool err_stop); +int tc_setup_cb_call(struct tcf_block *block, struct tcf_exts *exts, + enum tc_setup_type type, void *type_data, bool err_stop); enum tc_block_command { TC_BLOCK_BIND, diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index b16c79c47be5..cdfdc24b89cf 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -1206,10 +1206,25 @@ static int tc_exts_setup_cb_egdev_call(struct tcf_exts *exts, return ok_count; } -int tc_setup_cb_call(struct tcf_exts *exts, enum tc_setup_type type, - void *type_data, bool err_stop) +int tc_setup_cb_call(struct tcf_block *block, struct tcf_exts *exts, + enum tc_setup_type type, void *type_data, bool err_stop) { - return tc_exts_setup_cb_egdev_call(exts, type, type_data, err_stop); + int ok_count; + int ret; + + ret = tcf_block_cb_call(block, type, type_data, err_stop); + if (ret < 0) + return ret; + ok_count = ret; + + if (!exts) + return ok_count; + ret = tc_exts_setup_cb_egdev_call(exts, type, type_data, err_stop); + if (ret < 0) + return ret; + ok_count += ret; + + return ok_count; } EXPORT_SYMBOL(tc_setup_cb_call); diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index 5b7bb968d1d4..76b4e0a1c92f 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -201,6 +201,7 @@ static void fl_hw_destroy_filter(struct tcf_proto *tp, struct cls_fl_filter *f) { struct tc_cls_flower_offload cls_flower = {}; struct net_device *dev = tp->q->dev_queue->dev; + struct tcf_block *block = tp->chain->block; tc_cls_common_offload_init(&cls_flower.common, tp); cls_flower.command = TC_CLSFLOWER_DESTROY; @@ -209,7 +210,7 @@ static void fl_hw_destroy_filter(struct tcf_proto *tp, struct cls_fl_filter *f) if (tc_can_offload(dev)) dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSFLOWER, &cls_flower); - tc_setup_cb_call(&f->exts, TC_SETUP_CLSFLOWER, + tc_setup_cb_call(block, &f->exts, TC_SETUP_CLSFLOWER, &cls_flower, false); } @@ -220,6 +221,7 @@ static int fl_hw_replace_filter(struct tcf_proto *tp, { struct net_device *dev = tp->q->dev_queue->dev; struct tc_cls_flower_offload cls_flower = {}; + struct tcf_block *block = tp->chain->block; bool skip_sw = tc_skip_sw(f->flags); int err; @@ -242,7 +244,7 @@ static int fl_hw_replace_filter(struct tcf_proto *tp, } } - err = tc_setup_cb_call(&f->exts, TC_SETUP_CLSFLOWER, + err = tc_setup_cb_call(block, &f->exts, TC_SETUP_CLSFLOWER, &cls_flower, skip_sw); if (err < 0) { fl_hw_destroy_filter(tp, f); @@ -261,6 +263,7 @@ static void fl_hw_update_stats(struct tcf_proto *tp, struct cls_fl_filter *f) { struct tc_cls_flower_offload cls_flower = {}; struct net_device *dev = tp->q->dev_queue->dev; + struct tcf_block *block = tp->chain->block; tc_cls_common_offload_init(&cls_flower.common, tp); cls_flower.command = TC_CLSFLOWER_STATS; @@ -270,7 +273,7 @@ static void fl_hw_update_stats(struct tcf_proto *tp, struct cls_fl_filter *f) if (tc_can_offload(dev)) dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_CLSFLOWER, &cls_flower); - tc_setup_cb_call(&f->exts, TC_SETUP_CLSFLOWER, + tc_setup_cb_call(block, &f->exts, TC_SETUP_CLSFLOWER, &cls_flower, false); } -- cgit v1.2.3 From d58d31a118690b578897749feda48416ac10ca43 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 19 Oct 2017 15:50:47 +0200 Subject: net: sched: remove unused classid field from tc_cls_common_offload It is no longer used by the drivers, so remove it. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/pkt_cls.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include/net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index fcca5a9d9880..04caa246e747 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -561,7 +561,6 @@ struct tc_cls_common_offload { u32 chain_index; __be16 protocol; u32 prio; - u32 classid; }; static inline void @@ -571,7 +570,6 @@ tc_cls_common_offload_init(struct tc_cls_common_offload *cls_common, cls_common->chain_index = tp->chain->index; cls_common->protocol = tp->protocol; cls_common->prio = tp->prio; - cls_common->classid = tp->classid; } struct tc_cls_u32_knode { -- cgit v1.2.3 From fa71212e91811ac67014ad19d4fc3b3c3446ccf7 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Thu, 19 Oct 2017 15:50:48 +0200 Subject: net: sched: remove unused is_classid_clsact_ingress/egress helpers These helpers are no longer in use by drivers, so remove them. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/pkt_sched.h | 13 ------------- 1 file changed, 13 deletions(-) (limited to 'include/net') diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index 2d234af15f3e..b8ecafce4ba1 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -135,19 +135,6 @@ static inline unsigned int psched_mtu(const struct net_device *dev) return dev->mtu + dev->hard_header_len; } -static inline bool is_classid_clsact_ingress(u32 classid) -{ - /* This also returns true for ingress qdisc */ - return TC_H_MAJ(classid) == TC_H_MAJ(TC_H_CLSACT) && - TC_H_MIN(classid) != TC_H_MIN(TC_H_MIN_EGRESS); -} - -static inline bool is_classid_clsact_egress(u32 classid) -{ - return TC_H_MAJ(classid) == TC_H_MAJ(TC_H_CLSACT) && - TC_H_MIN(classid) == TC_H_MIN(TC_H_MIN_EGRESS); -} - static inline struct net *qdisc_net(struct Qdisc *q) { return dev_net(q->dev_queue->dev); -- cgit v1.2.3 From 3f27fb23219e75343b094366f2358bff34300493 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 23 Oct 2017 16:17:47 -0700 Subject: ipv6: addrconf: add per netns perturbation in inet6_addr_hash() Bring IPv6 in par with IPv4 : - Use net_hash_mix() to spread addresses a bit more. - Use 256 slots hash table instead of 16 Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/addrconf.h | 2 +- net/ipv6/addrconf.c | 16 +++++++++------- 2 files changed, 10 insertions(+), 8 deletions(-) (limited to 'include/net') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index b8b16437c6d5..15b5ffd7253d 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -58,7 +58,7 @@ struct in6_validator_info { struct netlink_ext_ack *extack; }; -#define IN6_ADDR_HSIZE_SHIFT 4 +#define IN6_ADDR_HSIZE_SHIFT 8 #define IN6_ADDR_HSIZE (1 << IN6_ADDR_HSIZE_SHIFT) int addrconf_init(void); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index c1a5028f394c..d70d98122053 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -950,9 +950,11 @@ ipv6_link_dev_addr(struct inet6_dev *idev, struct inet6_ifaddr *ifp) list_add_tail_rcu(&ifp->if_list, p); } -static u32 inet6_addr_hash(const struct in6_addr *addr) +static u32 inet6_addr_hash(const struct net *net, const struct in6_addr *addr) { - return hash_32(ipv6_addr_hash(addr), IN6_ADDR_HSIZE_SHIFT); + u32 val = ipv6_addr_hash(addr) ^ net_hash_mix(net); + + return hash_32(val, IN6_ADDR_HSIZE_SHIFT); } static bool ipv6_chk_same_addr(struct net *net, const struct in6_addr *addr, @@ -973,7 +975,7 @@ static bool ipv6_chk_same_addr(struct net *net, const struct in6_addr *addr, static int ipv6_add_addr_hash(struct net_device *dev, struct inet6_ifaddr *ifa) { - unsigned int hash = inet6_addr_hash(&ifa->addr); + unsigned int hash = inet6_addr_hash(dev_net(dev), &ifa->addr); int err = 0; spin_lock(&addrconf_hash_lock); @@ -1838,8 +1840,8 @@ int ipv6_chk_addr_and_flags(struct net *net, const struct in6_addr *addr, const struct net_device *dev, int strict, u32 banned_flags) { + unsigned int hash = inet6_addr_hash(net, addr); struct inet6_ifaddr *ifp; - unsigned int hash = inet6_addr_hash(addr); u32 ifp_flags; rcu_read_lock_bh(); @@ -1917,8 +1919,8 @@ EXPORT_SYMBOL(ipv6_chk_prefix); struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, const struct in6_addr *addr, struct net_device *dev, int strict) { + unsigned int hash = inet6_addr_hash(net, addr); struct inet6_ifaddr *ifp, *result = NULL; - unsigned int hash = inet6_addr_hash(addr); rcu_read_lock_bh(); hlist_for_each_entry_rcu_bh(ifp, &inet6_addr_lst[hash], addr_lst) { @@ -4242,9 +4244,9 @@ void if6_proc_exit(void) /* Check if address is a home address configured on any interface. */ int ipv6_chk_home_addr(struct net *net, const struct in6_addr *addr) { - int ret = 0; + unsigned int hash = inet6_addr_hash(net, addr); struct inet6_ifaddr *ifp = NULL; - unsigned int hash = inet6_addr_hash(addr); + int ret = 0; rcu_read_lock_bh(); hlist_for_each_entry_rcu_bh(ifp, &inet6_addr_lst[hash], addr_lst) { -- cgit v1.2.3 From b6f4f8484d88b69f700907200a9a9ec73806355f Mon Sep 17 00:00:00 2001 From: Tim Hansen Date: Mon, 23 Oct 2017 15:35:58 -0400 Subject: net/sock: Update sk rcu iterator macro. Mark hlist node in sk rcu iterator as protected by the rcu. hlist_next_rcu accomplishes this and silences the warnings sparse throws. Found with make C=1 net/ipv4/udp.o on linux-next tag next-20171009. Signed-off-by: Tim Hansen Signed-off-by: David S. Miller --- include/net/sock.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index 4827094f1db4..6f1be9726e02 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -737,10 +737,10 @@ static inline void sk_add_bind_node(struct sock *sk, * */ #define sk_for_each_entry_offset_rcu(tpos, pos, head, offset) \ - for (pos = rcu_dereference((head)->first); \ + for (pos = rcu_dereference(hlist_first_rcu(head)); \ pos != NULL && \ ({ tpos = (typeof(*tpos) *)((void *)pos - offset); 1;}); \ - pos = rcu_dereference(pos->next)) + pos = rcu_dereference(hlist_next_rcu(pos))) static inline struct user_namespace *sk_user_ns(struct sock *sk) { -- cgit v1.2.3 From 71c02379c762cb616c00fd5c4ed253fbf6bbe11b Mon Sep 17 00:00:00 2001 From: Christoph Paasch Date: Mon, 23 Oct 2017 13:22:23 -0700 Subject: tcp: Configure TFO without cookie per socket and/or per route We already allow to enable TFO without a cookie by using the fastopen-sysctl and setting it to TFO_SERVER_COOKIE_NOT_REQD (or TFO_CLIENT_NO_COOKIE). This is safe to do in certain environments where we know that there isn't a malicous host (aka., data-centers) or when the application-protocol already provides an authentication mechanism in the first flight of data. A server however might be providing multiple services or talking to both sides (public Internet and data-center). So, this server would want to enable cookie-less TFO for certain services and/or for connections that go to the data-center. This patch exposes a socket-option and a per-route attribute to enable such fine-grained configurations. Signed-off-by: Christoph Paasch Reviewed-by: Yuchung Cheng Signed-off-by: David S. Miller --- include/linux/tcp.h | 3 ++- include/net/tcp.h | 3 ++- include/uapi/linux/rtnetlink.h | 2 ++ include/uapi/linux/tcp.h | 1 + net/ipv4/tcp.c | 12 ++++++++++++ net/ipv4/tcp_fastopen.c | 20 +++++++++++++++++--- net/ipv4/tcp_input.c | 2 +- 7 files changed, 37 insertions(+), 6 deletions(-) (limited to 'include/net') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 1d2c44e09e31..173a7c2f9636 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -215,7 +215,8 @@ struct tcp_sock { u8 chrono_type:2, /* current chronograph type */ rate_app_limited:1, /* rate_{delivered,interval_us} limited? */ fastopen_connect:1, /* FASTOPEN_CONNECT sockopt */ - unused:4; + fastopen_no_cookie:1, /* Allow send/recv SYN+data without a cookie */ + unused:3; u8 nonagle : 4,/* Disable Nagle algorithm? */ thin_lto : 1,/* Use linear timeouts for thin streams */ unused1 : 1, diff --git a/include/net/tcp.h b/include/net/tcp.h index 2c13484704cb..2392f74074e7 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1567,7 +1567,8 @@ int tcp_fastopen_reset_cipher(struct net *net, struct sock *sk, void tcp_fastopen_add_skb(struct sock *sk, struct sk_buff *skb); struct sock *tcp_try_fastopen(struct sock *sk, struct sk_buff *skb, struct request_sock *req, - struct tcp_fastopen_cookie *foc); + struct tcp_fastopen_cookie *foc, + const struct dst_entry *dst); void tcp_fastopen_init_key_once(struct net *net); bool tcp_fastopen_cookie_check(struct sock *sk, u16 *mss, struct tcp_fastopen_cookie *cookie); diff --git a/include/uapi/linux/rtnetlink.h b/include/uapi/linux/rtnetlink.h index dab7dad9e01a..fe6679268901 100644 --- a/include/uapi/linux/rtnetlink.h +++ b/include/uapi/linux/rtnetlink.h @@ -430,6 +430,8 @@ enum { #define RTAX_QUICKACK RTAX_QUICKACK RTAX_CC_ALGO, #define RTAX_CC_ALGO RTAX_CC_ALGO + RTAX_FASTOPEN_NO_COOKIE, +#define RTAX_FASTOPEN_NO_COOKIE RTAX_FASTOPEN_NO_COOKIE __RTAX_MAX }; diff --git a/include/uapi/linux/tcp.h b/include/uapi/linux/tcp.h index 69c7493e42f8..d67e1d40c6d6 100644 --- a/include/uapi/linux/tcp.h +++ b/include/uapi/linux/tcp.h @@ -120,6 +120,7 @@ enum { #define TCP_ULP 31 /* Attach a ULP to a TCP connection */ #define TCP_MD5SIG_EXT 32 /* TCP MD5 Signature with extensions */ #define TCP_FASTOPEN_KEY 33 /* Set the key for Fast Open (cookie) */ +#define TCP_FASTOPEN_NO_COOKIE 34 /* Enable TFO without a TFO cookie */ struct tcp_repair_opt { __u32 opt_code; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index be07e9b6dbdd..8f36277e82e9 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2836,6 +2836,14 @@ static int do_tcp_setsockopt(struct sock *sk, int level, err = -EOPNOTSUPP; } break; + case TCP_FASTOPEN_NO_COOKIE: + if (val > 1 || val < 0) + err = -EINVAL; + else if (!((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN))) + err = -EINVAL; + else + tp->fastopen_no_cookie = val; + break; case TCP_TIMESTAMP: if (!tp->repair) err = -EPERM; @@ -3256,6 +3264,10 @@ static int do_tcp_getsockopt(struct sock *sk, int level, val = tp->fastopen_connect; break; + case TCP_FASTOPEN_NO_COOKIE: + val = tp->fastopen_no_cookie; + break; + case TCP_TIMESTAMP: val = tcp_time_stamp_raw() + tp->tsoffset; break; diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index 21075ce19cb6..e0a4b56644aa 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -310,13 +310,23 @@ static bool tcp_fastopen_queue_check(struct sock *sk) return true; } +static bool tcp_fastopen_no_cookie(const struct sock *sk, + const struct dst_entry *dst, + int flag) +{ + return (sock_net(sk)->ipv4.sysctl_tcp_fastopen & flag) || + tcp_sk(sk)->fastopen_no_cookie || + (dst && dst_metric(dst, RTAX_FASTOPEN_NO_COOKIE)); +} + /* Returns true if we should perform Fast Open on the SYN. The cookie (foc) * may be updated and return the client in the SYN-ACK later. E.g., Fast Open * cookie request (foc->len == 0). */ struct sock *tcp_try_fastopen(struct sock *sk, struct sk_buff *skb, struct request_sock *req, - struct tcp_fastopen_cookie *foc) + struct tcp_fastopen_cookie *foc, + const struct dst_entry *dst) { bool syn_data = TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq + 1; int tcp_fastopen = sock_net(sk)->ipv4.sysctl_tcp_fastopen; @@ -333,7 +343,8 @@ struct sock *tcp_try_fastopen(struct sock *sk, struct sk_buff *skb, return NULL; } - if (syn_data && (tcp_fastopen & TFO_SERVER_COOKIE_NOT_REQD)) + if (syn_data && + tcp_fastopen_no_cookie(sk, dst, TFO_SERVER_COOKIE_NOT_REQD)) goto fastopen; if (foc->len >= 0 && /* Client presents or requests a cookie */ @@ -370,6 +381,7 @@ bool tcp_fastopen_cookie_check(struct sock *sk, u16 *mss, struct tcp_fastopen_cookie *cookie) { unsigned long last_syn_loss = 0; + const struct dst_entry *dst; int syn_loss = 0; tcp_fastopen_cache_get(sk, mss, cookie, &syn_loss, &last_syn_loss); @@ -387,7 +399,9 @@ bool tcp_fastopen_cookie_check(struct sock *sk, u16 *mss, return false; } - if (sock_net(sk)->ipv4.sysctl_tcp_fastopen & TFO_CLIENT_NO_COOKIE) { + dst = __sk_dst_get(sk); + + if (tcp_fastopen_no_cookie(sk, dst, TFO_CLIENT_NO_COOKIE)) { cookie->len = -1; return true; } diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index c5e64d4b5839..893286db4623 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -6332,7 +6332,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, tcp_openreq_init_rwin(req, sk, dst); if (!want_cookie) { tcp_reqsk_record_syn(sk, req, skb); - fastopen_sk = tcp_try_fastopen(sk, skb, req, &foc); + fastopen_sk = tcp_try_fastopen(sk, skb, req, &foc, dst); } if (fastopen_sk) { af_ops->send_synack(fastopen_sk, dst, &fl, req, -- cgit v1.2.3 From c4f3db15958277c03d1c324894255ea3ecbf86e1 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 11 Oct 2017 10:47:40 +0200 Subject: netfilter: conntrack: add and use nf_l4proto_log_invalid We currently pass down the l4 protocol to the conntrack ->packet() function, but the only user of this is the debug info decision. Same information can be derived from struct nf_conn. As a first step, add and use a new log function for this, similar to nf_ct_helper_log(). Add __cold annotation -- invalid packets should be infrequent so gcc can consider all call paths that lead to such a function as unlikely. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_conntrack_l4proto.h | 10 +++++++ net/ipv4/netfilter/nf_conntrack_proto_icmp.c | 18 ++++++------ net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 14 +++++---- net/netfilter/nf_conntrack_proto.c | 24 ++++++++++++++++ net/netfilter/nf_conntrack_proto_dccp.c | 3 +- net/netfilter/nf_conntrack_proto_sctp.c | 3 +- net/netfilter/nf_conntrack_proto_tcp.c | 22 +++++++------- net/netfilter/nf_conntrack_proto_udp.c | 40 ++++++++++++-------------- 8 files changed, 82 insertions(+), 52 deletions(-) (limited to 'include/net') diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h index 738a0307a96b..6d79a061d360 100644 --- a/include/net/netfilter/nf_conntrack_l4proto.h +++ b/include/net/netfilter/nf_conntrack_l4proto.h @@ -152,8 +152,18 @@ extern const struct nla_policy nf_ct_port_nla_policy[]; #define LOG_INVALID(net, proto) \ ((net)->ct.sysctl_log_invalid == (proto) || \ (net)->ct.sysctl_log_invalid == IPPROTO_RAW) + +__printf(5, 6) __cold +void nf_l4proto_log_invalid(const struct sk_buff *skb, + struct net *net, + u16 pf, u8 protonum, + const char *fmt, ...); #else static inline int LOG_INVALID(struct net *net, int proto) { return 0; } + +static inline __printf(5, 6) __cold +void nf_l4proto_log_invalid(const struct sk_buff *skb, struct net *net, + u16 pf, u8 protonum, const char *fmt, ...) {} #endif /* CONFIG_SYSCTL */ #endif /*_NF_CONNTRACK_PROTOCOL_H*/ diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c index a046c298413a..7281a7b77a0e 100644 --- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c +++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c @@ -165,6 +165,12 @@ icmp_error_message(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb, return NF_ACCEPT; } +static void icmp_error_log(const struct sk_buff *skb, struct net *net, + u8 pf, const char *msg) +{ + nf_l4proto_log_invalid(skb, net, pf, IPPROTO_ICMP, "%s", msg); +} + /* Small and modified version of icmp_rcv */ static int icmp_error(struct net *net, struct nf_conn *tmpl, @@ -177,18 +183,14 @@ icmp_error(struct net *net, struct nf_conn *tmpl, /* Not enough header? */ icmph = skb_header_pointer(skb, ip_hdrlen(skb), sizeof(_ih), &_ih); if (icmph == NULL) { - if (LOG_INVALID(net, IPPROTO_ICMP)) - nf_log_packet(net, PF_INET, 0, skb, NULL, NULL, - NULL, "nf_ct_icmp: short packet "); + icmp_error_log(skb, net, pf, "short packet"); return -NF_ACCEPT; } /* See ip_conntrack_proto_tcp.c */ if (net->ct.sysctl_checksum && hooknum == NF_INET_PRE_ROUTING && nf_ip_checksum(skb, hooknum, dataoff, 0)) { - if (LOG_INVALID(net, IPPROTO_ICMP)) - nf_log_packet(net, PF_INET, 0, skb, NULL, NULL, NULL, - "nf_ct_icmp: bad HW ICMP checksum "); + icmp_error_log(skb, net, pf, "bad hw icmp checksum"); return -NF_ACCEPT; } @@ -199,9 +201,7 @@ icmp_error(struct net *net, struct nf_conn *tmpl, * discarded. */ if (icmph->type > NR_ICMP_TYPES) { - if (LOG_INVALID(net, IPPROTO_ICMP)) - nf_log_packet(net, PF_INET, 0, skb, NULL, NULL, NULL, - "nf_ct_icmp: invalid ICMP type "); + icmp_error_log(skb, net, pf, "invalid icmp type"); return -NF_ACCEPT; } diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c index a9e1fd1a8536..0f227ca4a5a2 100644 --- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c +++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c @@ -176,6 +176,12 @@ icmpv6_error_message(struct net *net, struct nf_conn *tmpl, return NF_ACCEPT; } +static void icmpv6_error_log(const struct sk_buff *skb, struct net *net, + u8 pf, const char *msg) +{ + nf_l4proto_log_invalid(skb, net, pf, IPPROTO_ICMPV6, "%s", msg); +} + static int icmpv6_error(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb, unsigned int dataoff, @@ -187,17 +193,13 @@ icmpv6_error(struct net *net, struct nf_conn *tmpl, icmp6h = skb_header_pointer(skb, dataoff, sizeof(_ih), &_ih); if (icmp6h == NULL) { - if (LOG_INVALID(net, IPPROTO_ICMPV6)) - nf_log_packet(net, PF_INET6, 0, skb, NULL, NULL, NULL, - "nf_ct_icmpv6: short packet "); + icmpv6_error_log(skb, net, pf, "short packet"); return -NF_ACCEPT; } if (net->ct.sysctl_checksum && hooknum == NF_INET_PRE_ROUTING && nf_ip6_checksum(skb, hooknum, dataoff, IPPROTO_ICMPV6)) { - if (LOG_INVALID(net, IPPROTO_ICMPV6)) - nf_log_packet(net, PF_INET6, 0, skb, NULL, NULL, NULL, - "nf_ct_icmpv6: ICMPv6 checksum failed "); + icmpv6_error_log(skb, net, pf, "ICMPv6 checksum failed"); return -NF_ACCEPT; } diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index b3e489c859ec..bcd3ee270d75 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -27,6 +27,7 @@ #include #include #include +#include static struct nf_conntrack_l4proto __rcu **nf_ct_protos[NFPROTO_NUMPROTO] __read_mostly; struct nf_conntrack_l3proto __rcu *nf_ct_l3protos[NFPROTO_NUMPROTO] __read_mostly; @@ -63,6 +64,29 @@ nf_ct_unregister_sysctl(struct ctl_table_header **header, *header = NULL; *table = NULL; } + +__printf(5, 6) +void nf_l4proto_log_invalid(const struct sk_buff *skb, + struct net *net, + u16 pf, u8 protonum, + const char *fmt, ...) +{ + struct va_format vaf; + va_list args; + + if (net->ct.sysctl_log_invalid != protonum || + net->ct.sysctl_log_invalid != IPPROTO_RAW) + return; + + va_start(args, fmt); + vaf.fmt = fmt; + vaf.va = &args; + + nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, + "nf_ct_proto_%d: %pV ", protonum, &vaf); + va_end(args); +} +EXPORT_SYMBOL_GPL(nf_l4proto_log_invalid); #endif const struct nf_conntrack_l4proto * diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c index 0f5a4d79f6b8..ef501c7edb96 100644 --- a/net/netfilter/nf_conntrack_proto_dccp.c +++ b/net/netfilter/nf_conntrack_proto_dccp.c @@ -604,8 +604,7 @@ static int dccp_error(struct net *net, struct nf_conn *tmpl, return NF_ACCEPT; out_invalid: - if (LOG_INVALID(net, IPPROTO_DCCP)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, "%s", msg); + nf_l4proto_log_invalid(skb, net, pf, IPPROTO_DCCP, "%s", msg); return -NF_ACCEPT; } diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c index 6303a88af12b..aa630c561361 100644 --- a/net/netfilter/nf_conntrack_proto_sctp.c +++ b/net/netfilter/nf_conntrack_proto_sctp.c @@ -522,8 +522,7 @@ static int sctp_error(struct net *net, struct nf_conn *tpl, struct sk_buff *skb, } return NF_ACCEPT; out_invalid: - if (LOG_INVALID(net, IPPROTO_SCTP)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, "%s", logmsg); + nf_l4proto_log_invalid(skb, net, pf, IPPROTO_SCTP, "%s", logmsg); return -NF_ACCEPT; } diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index cba1c6ffe51a..14198b2a2e2c 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -738,6 +738,12 @@ static const u8 tcp_valid_flags[(TCPHDR_FIN|TCPHDR_SYN|TCPHDR_RST|TCPHDR_ACK| [TCPHDR_ACK|TCPHDR_URG] = 1, }; +static void tcp_error_log(const struct sk_buff *skb, struct net *net, + u8 pf, const char *msg) +{ + nf_l4proto_log_invalid(skb, net, pf, IPPROTO_TCP, "%s", msg); +} + /* Protect conntrack agaist broken packets. Code taken from ipt_unclean.c. */ static int tcp_error(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb, @@ -753,17 +759,13 @@ static int tcp_error(struct net *net, struct nf_conn *tmpl, /* Smaller that minimal TCP header? */ th = skb_header_pointer(skb, dataoff, sizeof(_tcph), &_tcph); if (th == NULL) { - if (LOG_INVALID(net, IPPROTO_TCP)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_tcp: short packet "); + tcp_error_log(skb, net, pf, "short packet"); return -NF_ACCEPT; } /* Not whole TCP header or malformed packet */ if (th->doff*4 < sizeof(struct tcphdr) || tcplen < th->doff*4) { - if (LOG_INVALID(net, IPPROTO_TCP)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_tcp: truncated/malformed packet "); + tcp_error_log(skb, net, pf, "truncated packet"); return -NF_ACCEPT; } @@ -774,18 +776,14 @@ static int tcp_error(struct net *net, struct nf_conn *tmpl, /* FIXME: Source route IP option packets --RR */ if (net->ct.sysctl_checksum && hooknum == NF_INET_PRE_ROUTING && nf_checksum(skb, hooknum, dataoff, IPPROTO_TCP, pf)) { - if (LOG_INVALID(net, IPPROTO_TCP)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_tcp: bad TCP checksum "); + tcp_error_log(skb, net, pf, "bad checksum"); return -NF_ACCEPT; } /* Check TCP flags. */ tcpflags = (tcp_flag_byte(th) & ~(TCPHDR_ECE|TCPHDR_CWR|TCPHDR_PSH)); if (!tcp_valid_flags[tcpflags]) { - if (LOG_INVALID(net, IPPROTO_TCP)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_tcp: invalid TCP flag combination "); + tcp_error_log(skb, net, pf, "invalid tcp flag combination"); return -NF_ACCEPT; } diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c index 8af734cd1a94..fc20cf430251 100644 --- a/net/netfilter/nf_conntrack_proto_udp.c +++ b/net/netfilter/nf_conntrack_proto_udp.c @@ -99,6 +99,12 @@ static bool udp_new(struct nf_conn *ct, const struct sk_buff *skb, } #ifdef CONFIG_NF_CT_PROTO_UDPLITE +static void udplite_error_log(const struct sk_buff *skb, struct net *net, + u8 pf, const char *msg) +{ + nf_l4proto_log_invalid(skb, net, pf, IPPROTO_UDPLITE, "%s", msg); +} + static int udplite_error(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb, unsigned int dataoff, @@ -112,9 +118,7 @@ static int udplite_error(struct net *net, struct nf_conn *tmpl, /* Header is too small? */ hdr = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); if (!hdr) { - if (LOG_INVALID(net, IPPROTO_UDPLITE)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_udplite: short packet "); + udplite_error_log(skb, net, pf, "short packet"); return -NF_ACCEPT; } @@ -122,17 +126,13 @@ static int udplite_error(struct net *net, struct nf_conn *tmpl, if (cscov == 0) { cscov = udplen; } else if (cscov < sizeof(*hdr) || cscov > udplen) { - if (LOG_INVALID(net, IPPROTO_UDPLITE)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_udplite: invalid checksum coverage "); + udplite_error_log(skb, net, pf, "invalid checksum coverage"); return -NF_ACCEPT; } /* UDPLITE mandates checksums */ if (!hdr->check) { - if (LOG_INVALID(net, IPPROTO_UDPLITE)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_udplite: checksum missing "); + udplite_error_log(skb, net, pf, "checksum missing"); return -NF_ACCEPT; } @@ -140,9 +140,7 @@ static int udplite_error(struct net *net, struct nf_conn *tmpl, if (net->ct.sysctl_checksum && hooknum == NF_INET_PRE_ROUTING && nf_checksum_partial(skb, hooknum, dataoff, cscov, IPPROTO_UDP, pf)) { - if (LOG_INVALID(net, IPPROTO_UDPLITE)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_udplite: bad UDPLite checksum "); + udplite_error_log(skb, net, pf, "bad checksum"); return -NF_ACCEPT; } @@ -150,6 +148,12 @@ static int udplite_error(struct net *net, struct nf_conn *tmpl, } #endif +static void udp_error_log(const struct sk_buff *skb, struct net *net, + u8 pf, const char *msg) +{ + nf_l4proto_log_invalid(skb, net, pf, IPPROTO_UDP, "%s", msg); +} + static int udp_error(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb, unsigned int dataoff, u_int8_t pf, @@ -162,17 +166,13 @@ static int udp_error(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb, /* Header is too small? */ hdr = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); if (hdr == NULL) { - if (LOG_INVALID(net, IPPROTO_UDP)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_udp: short packet "); + udp_error_log(skb, net, pf, "short packet"); return -NF_ACCEPT; } /* Truncated/malformed packets */ if (ntohs(hdr->len) > udplen || ntohs(hdr->len) < sizeof(*hdr)) { - if (LOG_INVALID(net, IPPROTO_UDP)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_udp: truncated/malformed packet "); + udp_error_log(skb, net, pf, "truncated/malformed packet"); return -NF_ACCEPT; } @@ -186,9 +186,7 @@ static int udp_error(struct net *net, struct nf_conn *tmpl, struct sk_buff *skb, * FIXME: Source route IP option packets --RR */ if (net->ct.sysctl_checksum && hooknum == NF_INET_PRE_ROUTING && nf_checksum(skb, hooknum, dataoff, IPPROTO_UDP, pf)) { - if (LOG_INVALID(net, IPPROTO_UDP)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_udp: bad UDP checksum "); + udp_error_log(skb, net, pf, "bad checksum"); return -NF_ACCEPT; } -- cgit v1.2.3 From 3d0b527bc9dc0e8c4428eb1a98d4cd27bd1114c7 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 11 Oct 2017 10:47:41 +0200 Subject: netfilter: conntrack: add and use nf_ct_l4proto_log_invalid We currently pass down the l4 protocol to the conntrack ->packet() function, but the only user of this is the debug info decision. Same information can be derived from struct nf_conn. Add a wrapper for the previous patch that extracs the information from nf_conn and passes it to nf_l4proto_log_invalid(). Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_conntrack_l4proto.h | 14 ++++++++------ net/netfilter/nf_conntrack_proto.c | 23 +++++++++++++++++++++++ net/netfilter/nf_conntrack_proto_dccp.c | 17 +++++------------ net/netfilter/nf_conntrack_proto_tcp.c | 25 +++++++++---------------- 4 files changed, 45 insertions(+), 34 deletions(-) (limited to 'include/net') diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h index 6d79a061d360..5d51255b5bfb 100644 --- a/include/net/netfilter/nf_conntrack_l4proto.h +++ b/include/net/netfilter/nf_conntrack_l4proto.h @@ -149,21 +149,23 @@ int nf_ct_port_nlattr_tuple_size(void); extern const struct nla_policy nf_ct_port_nla_policy[]; #ifdef CONFIG_SYSCTL -#define LOG_INVALID(net, proto) \ - ((net)->ct.sysctl_log_invalid == (proto) || \ - (net)->ct.sysctl_log_invalid == IPPROTO_RAW) - +__printf(3, 4) __cold +void nf_ct_l4proto_log_invalid(const struct sk_buff *skb, + const struct nf_conn *ct, + const char *fmt, ...); __printf(5, 6) __cold void nf_l4proto_log_invalid(const struct sk_buff *skb, struct net *net, u16 pf, u8 protonum, const char *fmt, ...); #else -static inline int LOG_INVALID(struct net *net, int proto) { return 0; } - static inline __printf(5, 6) __cold void nf_l4proto_log_invalid(const struct sk_buff *skb, struct net *net, u16 pf, u8 protonum, const char *fmt, ...) {} +static inline __printf(3, 4) __cold +void nf_ct_l4proto_log_invalid(const struct sk_buff *skb, + const struct nf_conn *ct, + const char *fmt, ...) { } #endif /* CONFIG_SYSCTL */ #endif /*_NF_CONNTRACK_PROTOCOL_H*/ diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index bcd3ee270d75..83f739e9dc08 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -87,6 +87,29 @@ void nf_l4proto_log_invalid(const struct sk_buff *skb, va_end(args); } EXPORT_SYMBOL_GPL(nf_l4proto_log_invalid); + +__printf(3, 4) +void nf_ct_l4proto_log_invalid(const struct sk_buff *skb, + const struct nf_conn *ct, + const char *fmt, ...) +{ + struct va_format vaf; + struct net *net; + va_list args; + + net = nf_ct_net(ct); + if (likely(net->ct.sysctl_log_invalid == 0)) + return; + + va_start(args, fmt); + vaf.fmt = fmt; + vaf.va = &args; + + nf_l4proto_log_invalid(skb, net, nf_ct_l3num(ct), + nf_ct_protonum(ct), "%pV", &vaf); + va_end(args); +} +EXPORT_SYMBOL_GPL(nf_ct_l4proto_log_invalid); #endif const struct nf_conntrack_l4proto * diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c index ef501c7edb96..49e0abcdc6f4 100644 --- a/net/netfilter/nf_conntrack_proto_dccp.c +++ b/net/netfilter/nf_conntrack_proto_dccp.c @@ -428,13 +428,13 @@ static bool dccp_new(struct nf_conn *ct, const struct sk_buff *skb, default: dn = dccp_pernet(net); if (dn->dccp_loose == 0) { - msg = "nf_ct_dccp: not picking up existing connection "; + msg = "not picking up existing connection "; goto out_invalid; } case CT_DCCP_REQUEST: break; case CT_DCCP_INVALID: - msg = "nf_ct_dccp: invalid state transition "; + msg = "invalid state transition "; goto out_invalid; } @@ -447,9 +447,7 @@ static bool dccp_new(struct nf_conn *ct, const struct sk_buff *skb, return true; out_invalid: - if (LOG_INVALID(net, IPPROTO_DCCP)) - nf_log_packet(net, nf_ct_l3num(ct), 0, skb, NULL, NULL, - NULL, "%s", msg); + nf_ct_l4proto_log_invalid(skb, ct, "%s", msg); return false; } @@ -472,7 +470,6 @@ static int dccp_packet(struct nf_conn *ct, const struct sk_buff *skb, u_int8_t pf, unsigned int *timeouts) { - struct net *net = nf_ct_net(ct); enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); struct dccp_hdr _dh, *dh; u_int8_t type, old_state, new_state; @@ -534,15 +531,11 @@ static int dccp_packet(struct nf_conn *ct, const struct sk_buff *skb, ct->proto.dccp.last_pkt = type; spin_unlock_bh(&ct->lock); - if (LOG_INVALID(net, IPPROTO_DCCP)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_dccp: invalid packet ignored "); + nf_ct_l4proto_log_invalid(skb, ct, "%s", "invalid packet"); return NF_ACCEPT; case CT_DCCP_INVALID: spin_unlock_bh(&ct->lock); - if (LOG_INVALID(net, IPPROTO_DCCP)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_dccp: invalid state transition "); + nf_ct_l4proto_log_invalid(skb, ct, "%s", "invalid state transition"); return -NF_ACCEPT; } diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index 14198b2a2e2c..dced574f6006 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -702,9 +702,9 @@ static bool tcp_in_window(const struct nf_conn *ct, if (sender->flags & IP_CT_TCP_FLAG_BE_LIBERAL || tn->tcp_be_liberal) res = true; - if (!res && LOG_INVALID(net, IPPROTO_TCP)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_tcp: %s ", + if (!res) { + nf_ct_l4proto_log_invalid(skb, ct, + "%s", before(seq, sender->td_maxend + 1) ? in_recv_win ? before(sack, receiver->td_end + 1) ? @@ -713,6 +713,7 @@ static bool tcp_in_window(const struct nf_conn *ct, : "ACK is over the upper bound (ACKed data not seen yet)" : "SEQ is under the lower bound (already ACKed data retransmitted)" : "SEQ is over the upper bound (over the window of the receiver)"); + } } pr_debug("tcp_in_window: res=%u sender end=%u maxend=%u maxwin=%u " @@ -937,10 +938,8 @@ static int tcp_packet(struct nf_conn *ct, IP_CT_EXP_CHALLENGE_ACK; } spin_unlock_bh(&ct->lock); - if (LOG_INVALID(net, IPPROTO_TCP)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_tcp: invalid packet ignored in " - "state %s ", tcp_conntrack_names[old_state]); + nf_ct_l4proto_log_invalid(skb, ct, "invalid packet ignored in " + "state %s ", tcp_conntrack_names[old_state]); return NF_ACCEPT; case TCP_CONNTRACK_MAX: /* Special case for SYN proxy: when the SYN to the server or @@ -962,9 +961,7 @@ static int tcp_packet(struct nf_conn *ct, pr_debug("nf_ct_tcp: Invalid dir=%i index=%u ostate=%u\n", dir, get_conntrack_index(th), old_state); spin_unlock_bh(&ct->lock); - if (LOG_INVALID(net, IPPROTO_TCP)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_tcp: invalid state "); + nf_ct_l4proto_log_invalid(skb, ct, "invalid state"); return -NF_ACCEPT; case TCP_CONNTRACK_TIME_WAIT: /* RFC5961 compliance cause stack to send "challenge-ACK" @@ -979,9 +976,7 @@ static int tcp_packet(struct nf_conn *ct, /* Detected RFC5961 challenge ACK */ ct->proto.tcp.last_flags &= ~IP_CT_EXP_CHALLENGE_ACK; spin_unlock_bh(&ct->lock); - if (LOG_INVALID(net, IPPROTO_TCP)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, NULL, - "nf_ct_tcp: challenge-ACK ignored "); + nf_ct_l4proto_log_invalid(skb, ct, "challenge-ack ignored"); return NF_ACCEPT; /* Don't change state */ } break; @@ -991,9 +986,7 @@ static int tcp_packet(struct nf_conn *ct, && before(ntohl(th->seq), ct->proto.tcp.seen[!dir].td_maxack)) { /* Invalid RST */ spin_unlock_bh(&ct->lock); - if (LOG_INVALID(net, IPPROTO_TCP)) - nf_log_packet(net, pf, 0, skb, NULL, NULL, - NULL, "nf_ct_tcp: invalid RST "); + nf_ct_l4proto_log_invalid(skb, ct, "invalid rst"); return -NF_ACCEPT; } if (index == TCP_RST_SET -- cgit v1.2.3 From eb6fad5a4a328b85d3faa8b301b522e3f316b49d Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 11 Oct 2017 10:47:42 +0200 Subject: netfilter: conntrack: remove pf argument from l4 packet functions not needed/used anymore. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_conntrack_l4proto.h | 1 - net/ipv4/netfilter/nf_conntrack_proto_icmp.c | 1 - net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 1 - net/netfilter/nf_conntrack_core.c | 2 +- net/netfilter/nf_conntrack_proto_dccp.c | 1 - net/netfilter/nf_conntrack_proto_generic.c | 1 - net/netfilter/nf_conntrack_proto_gre.c | 1 - net/netfilter/nf_conntrack_proto_sctp.c | 1 - net/netfilter/nf_conntrack_proto_tcp.c | 6 ++---- net/netfilter/nf_conntrack_proto_udp.c | 1 - 10 files changed, 3 insertions(+), 13 deletions(-) (limited to 'include/net') diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h index 5d51255b5bfb..e06518874144 100644 --- a/include/net/netfilter/nf_conntrack_l4proto.h +++ b/include/net/netfilter/nf_conntrack_l4proto.h @@ -42,7 +42,6 @@ struct nf_conntrack_l4proto { const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - u_int8_t pf, unsigned int *timeouts); /* Called when a new connection for this protocol found; diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c index 7281a7b77a0e..8969420cecc3 100644 --- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c +++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c @@ -81,7 +81,6 @@ static int icmp_packet(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - u_int8_t pf, unsigned int *timeout) { /* Do not immediately delete the connection after the first diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c index 0f227ca4a5a2..dca921df28e1 100644 --- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c +++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c @@ -94,7 +94,6 @@ static int icmpv6_packet(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - u_int8_t pf, unsigned int *timeout) { /* Do not immediately delete the connection after the first diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 01130392b7c0..28e675150853 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -1419,7 +1419,7 @@ repeat: /* Decide what timeout policy we want to apply to this flow. */ timeouts = nf_ct_timeout_lookup(net, ct, l4proto); - ret = l4proto->packet(ct, skb, dataoff, ctinfo, pf, timeouts); + ret = l4proto->packet(ct, skb, dataoff, ctinfo, timeouts); if (ret <= 0) { /* Invalid: inverse of the return code tells * the netfilter core what to do */ diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c index 49e0abcdc6f4..2a446f4a554c 100644 --- a/net/netfilter/nf_conntrack_proto_dccp.c +++ b/net/netfilter/nf_conntrack_proto_dccp.c @@ -467,7 +467,6 @@ static unsigned int *dccp_get_timeouts(struct net *net) static int dccp_packet(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - u_int8_t pf, unsigned int *timeouts) { enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); diff --git a/net/netfilter/nf_conntrack_proto_generic.c b/net/netfilter/nf_conntrack_proto_generic.c index 9cd40700842e..1f86ddf6649a 100644 --- a/net/netfilter/nf_conntrack_proto_generic.c +++ b/net/netfilter/nf_conntrack_proto_generic.c @@ -60,7 +60,6 @@ static int generic_packet(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - u_int8_t pf, unsigned int *timeout) { nf_ct_refresh_acct(ct, ctinfo, skb, *timeout); diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c index 09a90484c27d..a2503005d80b 100644 --- a/net/netfilter/nf_conntrack_proto_gre.c +++ b/net/netfilter/nf_conntrack_proto_gre.c @@ -244,7 +244,6 @@ static int gre_packet(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - u_int8_t pf, unsigned int *timeouts) { /* If we've seen traffic both ways, this is a GRE connection. diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c index aa630c561361..80faf04ddf15 100644 --- a/net/netfilter/nf_conntrack_proto_sctp.c +++ b/net/netfilter/nf_conntrack_proto_sctp.c @@ -306,7 +306,6 @@ static int sctp_packet(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - u_int8_t pf, unsigned int *timeouts) { enum sctp_conntrack new_state, old_state; diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index dced574f6006..8f283294d70f 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -493,8 +493,7 @@ static bool tcp_in_window(const struct nf_conn *ct, unsigned int index, const struct sk_buff *skb, unsigned int dataoff, - const struct tcphdr *tcph, - u_int8_t pf) + const struct tcphdr *tcph) { struct net *net = nf_ct_net(ct); struct nf_tcp_net *tn = tcp_pernet(net); @@ -801,7 +800,6 @@ static int tcp_packet(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - u_int8_t pf, unsigned int *timeouts) { struct net *net = nf_ct_net(ct); @@ -1013,7 +1011,7 @@ static int tcp_packet(struct nf_conn *ct, } if (!tcp_in_window(ct, &ct->proto.tcp, dir, index, - skb, dataoff, th, pf)) { + skb, dataoff, th)) { spin_unlock_bh(&ct->lock); return -NF_ACCEPT; } diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c index fc20cf430251..3a5f727103af 100644 --- a/net/netfilter/nf_conntrack_proto_udp.c +++ b/net/netfilter/nf_conntrack_proto_udp.c @@ -73,7 +73,6 @@ static int udp_packet(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff, enum ip_conntrack_info ctinfo, - u_int8_t pf, unsigned int *timeouts) { /* If we've seen traffic both ways, this is some kind of UDP -- cgit v1.2.3 From 28efb0046512e8a13ed9f9bdf0d68d10bbfbe9cf Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 12 Oct 2017 09:38:30 +0200 Subject: netfilter: conntrack: make l3proto trackers const previous patches removed all writes to them. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/ipv4/nf_conntrack_ipv4.h | 2 +- include/net/netfilter/ipv6/nf_conntrack_ipv6.h | 2 +- net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 2 +- net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include/net') diff --git a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h index 919e4e8af327..5534ecca7a5d 100644 --- a/include/net/netfilter/ipv4/nf_conntrack_ipv4.h +++ b/include/net/netfilter/ipv4/nf_conntrack_ipv4.h @@ -10,7 +10,7 @@ #define _NF_CONNTRACK_IPV4_H -extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4; +const extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4; extern struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp4; extern struct nf_conntrack_l4proto nf_conntrack_l4proto_udp4; diff --git a/include/net/netfilter/ipv6/nf_conntrack_ipv6.h b/include/net/netfilter/ipv6/nf_conntrack_ipv6.h index eaea968f8657..30dc57980866 100644 --- a/include/net/netfilter/ipv6/nf_conntrack_ipv6.h +++ b/include/net/netfilter/ipv6/nf_conntrack_ipv6.h @@ -1,7 +1,7 @@ #ifndef _NF_CONNTRACK_IPV6_H #define _NF_CONNTRACK_IPV6_H -extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6; +extern const struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6; extern struct nf_conntrack_l4proto nf_conntrack_l4proto_tcp6; extern struct nf_conntrack_l4proto nf_conntrack_l4proto_udp6; diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index fe374da4bc13..89af9d88ca21 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -344,7 +344,7 @@ static void ipv4_hooks_unregister(struct net *net) mutex_unlock(®ister_ipv4_hooks); } -struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4 __read_mostly = { +const struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4 = { .l3proto = PF_INET, .pkt_to_tuple = ipv4_pkt_to_tuple, .invert_tuple = ipv4_invert_tuple, diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c index fe01dc953c56..3b80a38f62b8 100644 --- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c +++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c @@ -339,7 +339,7 @@ static void ipv6_hooks_unregister(struct net *net) mutex_unlock(®ister_ipv6_hooks); } -struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6 __read_mostly = { +const struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6 = { .l3proto = PF_INET6, .pkt_to_tuple = ipv6_pkt_to_tuple, .invert_tuple = ipv6_invert_tuple, -- cgit v1.2.3 From ef5201c83d1400570a3b6f004ad7a23d71934411 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Tue, 24 Oct 2017 13:54:20 +0800 Subject: bonding: remove rtmsg_ifinfo called after bond_lower_state_changed After the patch 'rtnetlink: bring NETDEV_CHANGELOWERSTATE event process back to rtnetlink_event', bond_lower_state_changed would generate NETDEV_CHANGEUPPER event which would send a notification to userspace in rtnetlink_event. There's no need to call rtmsg_ifinfo to send the notification any more. So this patch is to remove it from these places after bond_lower_state_changed. Besides, after this, rtmsg_ifinfo is not needed to be exported. Signed-off-by: Xin Long Signed-off-by: David S. Miller --- include/net/bonding.h | 4 ---- net/core/rtnetlink.c | 1 - 2 files changed, 5 deletions(-) (limited to 'include/net') diff --git a/include/net/bonding.h b/include/net/bonding.h index 2860cc66c2bb..f801fc940b29 100644 --- a/include/net/bonding.h +++ b/include/net/bonding.h @@ -330,7 +330,6 @@ static inline void bond_set_active_slave(struct slave *slave) slave->backup = 0; bond_queue_slave_event(slave); bond_lower_state_changed(slave); - rtmsg_ifinfo(RTM_NEWLINK, slave->dev, 0, GFP_ATOMIC); } } @@ -340,7 +339,6 @@ static inline void bond_set_backup_slave(struct slave *slave) slave->backup = 1; bond_queue_slave_event(slave); bond_lower_state_changed(slave); - rtmsg_ifinfo(RTM_NEWLINK, slave->dev, 0, GFP_ATOMIC); } } @@ -353,7 +351,6 @@ static inline void bond_set_slave_state(struct slave *slave, slave->backup = slave_state; if (notify) { bond_lower_state_changed(slave); - rtmsg_ifinfo(RTM_NEWLINK, slave->dev, 0, GFP_ATOMIC); bond_queue_slave_event(slave); slave->should_notify = 0; } else { @@ -385,7 +382,6 @@ static inline void bond_slave_state_notify(struct bonding *bond) bond_for_each_slave(bond, tmp, iter) { if (tmp->should_notify) { bond_lower_state_changed(tmp); - rtmsg_ifinfo(RTM_NEWLINK, tmp->dev, 0, GFP_ATOMIC); tmp->should_notify = 0; } } diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 854a848842ea..de24d394c69e 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -2989,7 +2989,6 @@ void rtmsg_ifinfo(int type, struct net_device *dev, unsigned int change, { rtmsg_ifinfo_event(type, dev, change, rtnl_get_event(0), flags, NULL); } -EXPORT_SYMBOL(rtmsg_ifinfo); void rtmsg_ifinfo_newnet(int type, struct net_device *dev, unsigned int change, gfp_t flags, int *new_nsid) -- cgit v1.2.3 From 9c3b57518363577d4e2ea1964ef4fa03e100beaa Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Tue, 24 Oct 2017 01:45:31 -0700 Subject: net: sctp: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: Vlad Yasevich Cc: Neil Horman Cc: "David S. Miller" Cc: linux-sctp@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook Signed-off-by: David S. Miller --- include/net/sctp/sm.h | 10 +++--- net/sctp/associola.c | 3 +- net/sctp/protocol.c | 7 ++-- net/sctp/sm_sideeffect.c | 85 +++++++++++++++++++++++++++++------------------- net/sctp/transport.c | 13 +++----- 5 files changed, 66 insertions(+), 52 deletions(-) (limited to 'include/net') diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h index 2db3d3a9ce1d..13cc4963e905 100644 --- a/include/net/sctp/sm.h +++ b/include/net/sctp/sm.h @@ -72,7 +72,7 @@ typedef enum sctp_disposition (sctp_state_fn_t) ( const union sctp_subtype type, void *arg, struct sctp_cmd_seq *commands); -typedef void (sctp_timer_event_t) (unsigned long); +typedef void (sctp_timer_event_t) (struct timer_list *); struct sctp_sm_table_entry { sctp_state_fn_t *fn; const char *name; @@ -314,10 +314,10 @@ int sctp_do_sm(struct net *net, enum sctp_event event_type, void *event_arg, gfp_t gfp); /* 2nd level prototypes */ -void sctp_generate_t3_rtx_event(unsigned long peer); -void sctp_generate_heartbeat_event(unsigned long peer); -void sctp_generate_reconf_event(unsigned long peer); -void sctp_generate_proto_unreach_event(unsigned long peer); +void sctp_generate_t3_rtx_event(struct timer_list *t); +void sctp_generate_heartbeat_event(struct timer_list *t); +void sctp_generate_reconf_event(struct timer_list *t); +void sctp_generate_proto_unreach_event(struct timer_list *t); void sctp_ootb_pkt_free(struct sctp_packet *packet); diff --git a/net/sctp/associola.c b/net/sctp/associola.c index dfb9651e818b..69394f4d6091 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -149,8 +149,7 @@ static struct sctp_association *sctp_association_init( /* Initializes the timers */ for (i = SCTP_EVENT_TIMEOUT_NONE; i < SCTP_NUM_TIMEOUT_TYPES; ++i) - setup_timer(&asoc->timers[i], sctp_timer_events[i], - (unsigned long)asoc); + timer_setup(&asoc->timers[i], sctp_timer_events[i], 0); /* Pull default initialization values from the sock options. * Note: This assumes that the values have already been diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index fcd80feb293f..f5172c21349b 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -622,9 +622,9 @@ static void sctp_v4_ecn_capable(struct sock *sk) INET_ECN_xmit(sk); } -static void sctp_addr_wq_timeout_handler(unsigned long arg) +static void sctp_addr_wq_timeout_handler(struct timer_list *t) { - struct net *net = (struct net *)arg; + struct net *net = from_timer(net, t, sctp.addr_wq_timer); struct sctp_sockaddr_entry *addrw, *temp; struct sctp_sock *sp; @@ -1304,8 +1304,7 @@ static int __net_init sctp_defaults_init(struct net *net) INIT_LIST_HEAD(&net->sctp.auto_asconf_splist); spin_lock_init(&net->sctp.addr_wq_lock); net->sctp.addr_wq_timer.expires = 0; - setup_timer(&net->sctp.addr_wq_timer, sctp_addr_wq_timeout_handler, - (unsigned long)net); + timer_setup(&net->sctp.addr_wq_timer, sctp_addr_wq_timeout_handler, 0); return 0; diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index 402bfbb888cd..1c2699b424af 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -243,9 +243,10 @@ nomem: /* When the T3-RTX timer expires, it calls this function to create the * relevant state machine event. */ -void sctp_generate_t3_rtx_event(unsigned long peer) +void sctp_generate_t3_rtx_event(struct timer_list *t) { - struct sctp_transport *transport = (struct sctp_transport *) peer; + struct sctp_transport *transport = + from_timer(transport, t, T3_rtx_timer); struct sctp_association *asoc = transport->asoc; struct sock *sk = asoc->base.sk; struct net *net = sock_net(sk); @@ -319,50 +320,63 @@ out_unlock: sctp_association_put(asoc); } -static void sctp_generate_t1_cookie_event(unsigned long data) +static void sctp_generate_t1_cookie_event(struct timer_list *t) { - struct sctp_association *asoc = (struct sctp_association *) data; + struct sctp_association *asoc = + from_timer(asoc, t, timers[SCTP_EVENT_TIMEOUT_T1_COOKIE]); + sctp_generate_timeout_event(asoc, SCTP_EVENT_TIMEOUT_T1_COOKIE); } -static void sctp_generate_t1_init_event(unsigned long data) +static void sctp_generate_t1_init_event(struct timer_list *t) { - struct sctp_association *asoc = (struct sctp_association *) data; + struct sctp_association *asoc = + from_timer(asoc, t, timers[SCTP_EVENT_TIMEOUT_T1_INIT]); + sctp_generate_timeout_event(asoc, SCTP_EVENT_TIMEOUT_T1_INIT); } -static void sctp_generate_t2_shutdown_event(unsigned long data) +static void sctp_generate_t2_shutdown_event(struct timer_list *t) { - struct sctp_association *asoc = (struct sctp_association *) data; + struct sctp_association *asoc = + from_timer(asoc, t, timers[SCTP_EVENT_TIMEOUT_T2_SHUTDOWN]); + sctp_generate_timeout_event(asoc, SCTP_EVENT_TIMEOUT_T2_SHUTDOWN); } -static void sctp_generate_t4_rto_event(unsigned long data) +static void sctp_generate_t4_rto_event(struct timer_list *t) { - struct sctp_association *asoc = (struct sctp_association *) data; + struct sctp_association *asoc = + from_timer(asoc, t, timers[SCTP_EVENT_TIMEOUT_T4_RTO]); + sctp_generate_timeout_event(asoc, SCTP_EVENT_TIMEOUT_T4_RTO); } -static void sctp_generate_t5_shutdown_guard_event(unsigned long data) +static void sctp_generate_t5_shutdown_guard_event(struct timer_list *t) { - struct sctp_association *asoc = (struct sctp_association *)data; + struct sctp_association *asoc = + from_timer(asoc, t, + timers[SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD]); + sctp_generate_timeout_event(asoc, SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD); } /* sctp_generate_t5_shutdown_guard_event() */ -static void sctp_generate_autoclose_event(unsigned long data) +static void sctp_generate_autoclose_event(struct timer_list *t) { - struct sctp_association *asoc = (struct sctp_association *) data; + struct sctp_association *asoc = + from_timer(asoc, t, timers[SCTP_EVENT_TIMEOUT_AUTOCLOSE]); + sctp_generate_timeout_event(asoc, SCTP_EVENT_TIMEOUT_AUTOCLOSE); } /* Generate a heart beat event. If the sock is busy, reschedule. Make * sure that the transport is still valid. */ -void sctp_generate_heartbeat_event(unsigned long data) +void sctp_generate_heartbeat_event(struct timer_list *t) { - struct sctp_transport *transport = (struct sctp_transport *) data; + struct sctp_transport *transport = from_timer(transport, t, hb_timer); struct sctp_association *asoc = transport->asoc; struct sock *sk = asoc->base.sk; struct net *net = sock_net(sk); @@ -405,9 +419,10 @@ out_unlock: /* Handle the timeout of the ICMP protocol unreachable timer. Trigger * the correct state machine transition that will close the association. */ -void sctp_generate_proto_unreach_event(unsigned long data) +void sctp_generate_proto_unreach_event(struct timer_list *t) { - struct sctp_transport *transport = (struct sctp_transport *)data; + struct sctp_transport *transport = + from_timer(transport, t, proto_unreach_timer); struct sctp_association *asoc = transport->asoc; struct sock *sk = asoc->base.sk; struct net *net = sock_net(sk); @@ -439,9 +454,10 @@ out_unlock: } /* Handle the timeout of the RE-CONFIG timer. */ -void sctp_generate_reconf_event(unsigned long data) +void sctp_generate_reconf_event(struct timer_list *t) { - struct sctp_transport *transport = (struct sctp_transport *)data; + struct sctp_transport *transport = + from_timer(transport, t, reconf_timer); struct sctp_association *asoc = transport->asoc; struct sock *sk = asoc->base.sk; struct net *net = sock_net(sk); @@ -471,24 +487,27 @@ out_unlock: } /* Inject a SACK Timeout event into the state machine. */ -static void sctp_generate_sack_event(unsigned long data) +static void sctp_generate_sack_event(struct timer_list *t) { - struct sctp_association *asoc = (struct sctp_association *)data; + struct sctp_association *asoc = + from_timer(asoc, t, timers[SCTP_EVENT_TIMEOUT_SACK]); + sctp_generate_timeout_event(asoc, SCTP_EVENT_TIMEOUT_SACK); } sctp_timer_event_t *sctp_timer_events[SCTP_NUM_TIMEOUT_TYPES] = { - NULL, - sctp_generate_t1_cookie_event, - sctp_generate_t1_init_event, - sctp_generate_t2_shutdown_event, - NULL, - sctp_generate_t4_rto_event, - sctp_generate_t5_shutdown_guard_event, - NULL, - NULL, - sctp_generate_sack_event, - sctp_generate_autoclose_event, + [SCTP_EVENT_TIMEOUT_NONE] = NULL, + [SCTP_EVENT_TIMEOUT_T1_COOKIE] = sctp_generate_t1_cookie_event, + [SCTP_EVENT_TIMEOUT_T1_INIT] = sctp_generate_t1_init_event, + [SCTP_EVENT_TIMEOUT_T2_SHUTDOWN] = sctp_generate_t2_shutdown_event, + [SCTP_EVENT_TIMEOUT_T3_RTX] = NULL, + [SCTP_EVENT_TIMEOUT_T4_RTO] = sctp_generate_t4_rto_event, + [SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD] = + sctp_generate_t5_shutdown_guard_event, + [SCTP_EVENT_TIMEOUT_HEARTBEAT] = NULL, + [SCTP_EVENT_TIMEOUT_RECONF] = NULL, + [SCTP_EVENT_TIMEOUT_SACK] = sctp_generate_sack_event, + [SCTP_EVENT_TIMEOUT_AUTOCLOSE] = sctp_generate_autoclose_event, }; diff --git a/net/sctp/transport.c b/net/sctp/transport.c index 2d9bd3776bc8..1e5a22430cf5 100644 --- a/net/sctp/transport.c +++ b/net/sctp/transport.c @@ -87,14 +87,11 @@ static struct sctp_transport *sctp_transport_init(struct net *net, INIT_LIST_HEAD(&peer->send_ready); INIT_LIST_HEAD(&peer->transports); - setup_timer(&peer->T3_rtx_timer, sctp_generate_t3_rtx_event, - (unsigned long)peer); - setup_timer(&peer->hb_timer, sctp_generate_heartbeat_event, - (unsigned long)peer); - setup_timer(&peer->reconf_timer, sctp_generate_reconf_event, - (unsigned long)peer); - setup_timer(&peer->proto_unreach_timer, - sctp_generate_proto_unreach_event, (unsigned long)peer); + timer_setup(&peer->T3_rtx_timer, sctp_generate_t3_rtx_event, 0); + timer_setup(&peer->hb_timer, sctp_generate_heartbeat_event, 0); + timer_setup(&peer->reconf_timer, sctp_generate_reconf_event, 0); + timer_setup(&peer->proto_unreach_timer, + sctp_generate_proto_unreach_event, 0); /* Initialize the 64-bit random nonce sent with heartbeat. */ get_random_bytes(&peer->hb_nonce, sizeof(peer->hb_nonce)); -- cgit v1.2.3 From fc8bcaa05160528d56432e4612f522e3ceafc513 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Tue, 24 Oct 2017 01:45:48 -0700 Subject: net: LLC: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Cc: "David S. Miller" Cc: Eric Dumazet Cc: Hans Liljestrand Cc: "Paul E. McKenney" Cc: "Reshetova, Elena" Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook Signed-off-by: David S. Miller --- include/net/llc_c_ac.h | 8 ++++---- net/llc/llc_c_ac.c | 27 +++++++++++++++++---------- net/llc/llc_conn.c | 12 ++++-------- 3 files changed, 25 insertions(+), 22 deletions(-) (limited to 'include/net') diff --git a/include/net/llc_c_ac.h b/include/net/llc_c_ac.h index f3be818e73c1..e766300b3e99 100644 --- a/include/net/llc_c_ac.h +++ b/include/net/llc_c_ac.h @@ -171,10 +171,10 @@ int llc_conn_ac_rst_sendack_flag(struct sock *sk, struct sk_buff *skb); int llc_conn_ac_send_i_rsp_as_ack(struct sock *sk, struct sk_buff *skb); int llc_conn_ac_send_i_as_ack(struct sock *sk, struct sk_buff *skb); -void llc_conn_busy_tmr_cb(unsigned long timeout_data); -void llc_conn_pf_cycle_tmr_cb(unsigned long timeout_data); -void llc_conn_ack_tmr_cb(unsigned long timeout_data); -void llc_conn_rej_tmr_cb(unsigned long timeout_data); +void llc_conn_busy_tmr_cb(struct timer_list *t); +void llc_conn_pf_cycle_tmr_cb(struct timer_list *t); +void llc_conn_ack_tmr_cb(struct timer_list *t); +void llc_conn_rej_tmr_cb(struct timer_list *t); void llc_conn_set_p_flag(struct sock *sk, u8 value); #endif /* LLC_C_AC_H */ diff --git a/net/llc/llc_c_ac.c b/net/llc/llc_c_ac.c index ea225bd2672c..f59648018060 100644 --- a/net/llc/llc_c_ac.c +++ b/net/llc/llc_c_ac.c @@ -1318,9 +1318,8 @@ static int llc_conn_ac_inc_vs_by_1(struct sock *sk, struct sk_buff *skb) return 0; } -static void llc_conn_tmr_common_cb(unsigned long timeout_data, u8 type) +static void llc_conn_tmr_common_cb(struct sock *sk, u8 type) { - struct sock *sk = (struct sock *)timeout_data; struct sk_buff *skb = alloc_skb(0, GFP_ATOMIC); bh_lock_sock(sk); @@ -1334,24 +1333,32 @@ static void llc_conn_tmr_common_cb(unsigned long timeout_data, u8 type) bh_unlock_sock(sk); } -void llc_conn_pf_cycle_tmr_cb(unsigned long timeout_data) +void llc_conn_pf_cycle_tmr_cb(struct timer_list *t) { - llc_conn_tmr_common_cb(timeout_data, LLC_CONN_EV_TYPE_P_TMR); + struct llc_sock *llc = from_timer(llc, t, pf_cycle_timer.timer); + + llc_conn_tmr_common_cb(&llc->sk, LLC_CONN_EV_TYPE_P_TMR); } -void llc_conn_busy_tmr_cb(unsigned long timeout_data) +void llc_conn_busy_tmr_cb(struct timer_list *t) { - llc_conn_tmr_common_cb(timeout_data, LLC_CONN_EV_TYPE_BUSY_TMR); + struct llc_sock *llc = from_timer(llc, t, busy_state_timer.timer); + + llc_conn_tmr_common_cb(&llc->sk, LLC_CONN_EV_TYPE_BUSY_TMR); } -void llc_conn_ack_tmr_cb(unsigned long timeout_data) +void llc_conn_ack_tmr_cb(struct timer_list *t) { - llc_conn_tmr_common_cb(timeout_data, LLC_CONN_EV_TYPE_ACK_TMR); + struct llc_sock *llc = from_timer(llc, t, ack_timer.timer); + + llc_conn_tmr_common_cb(&llc->sk, LLC_CONN_EV_TYPE_ACK_TMR); } -void llc_conn_rej_tmr_cb(unsigned long timeout_data) +void llc_conn_rej_tmr_cb(struct timer_list *t) { - llc_conn_tmr_common_cb(timeout_data, LLC_CONN_EV_TYPE_REJ_TMR); + struct llc_sock *llc = from_timer(llc, t, rej_sent_timer.timer); + + llc_conn_tmr_common_cb(&llc->sk, LLC_CONN_EV_TYPE_REJ_TMR); } int llc_conn_ac_rst_vs(struct sock *sk, struct sk_buff *skb) diff --git a/net/llc/llc_conn.c b/net/llc/llc_conn.c index 5e91b47f0d2a..9177dbb16dce 100644 --- a/net/llc/llc_conn.c +++ b/net/llc/llc_conn.c @@ -902,20 +902,16 @@ static void llc_sk_init(struct sock *sk) llc->inc_cntr = llc->dec_cntr = 2; llc->dec_step = llc->connect_step = 1; - setup_timer(&llc->ack_timer.timer, llc_conn_ack_tmr_cb, - (unsigned long)sk); + timer_setup(&llc->ack_timer.timer, llc_conn_ack_tmr_cb, 0); llc->ack_timer.expire = sysctl_llc2_ack_timeout; - setup_timer(&llc->pf_cycle_timer.timer, llc_conn_pf_cycle_tmr_cb, - (unsigned long)sk); + timer_setup(&llc->pf_cycle_timer.timer, llc_conn_pf_cycle_tmr_cb, 0); llc->pf_cycle_timer.expire = sysctl_llc2_p_timeout; - setup_timer(&llc->rej_sent_timer.timer, llc_conn_rej_tmr_cb, - (unsigned long)sk); + timer_setup(&llc->rej_sent_timer.timer, llc_conn_rej_tmr_cb, 0); llc->rej_sent_timer.expire = sysctl_llc2_rej_timeout; - setup_timer(&llc->busy_state_timer.timer, llc_conn_busy_tmr_cb, - (unsigned long)sk); + timer_setup(&llc->busy_state_timer.timer, llc_conn_busy_tmr_cb, 0); llc->busy_state_timer.expire = sysctl_llc2_busy_timeout; llc->n2 = 2; /* max retransmit */ -- cgit v1.2.3 From 60e2a7780793bae0debc275a9ccd57f7da0cf195 Mon Sep 17 00:00:00 2001 From: Ursula Braun Date: Wed, 25 Oct 2017 11:01:45 +0200 Subject: tcp: TCP experimental option for SMC The SMC protocol [1] relies on the use of a new TCP experimental option [2, 3]. With this option, SMC capabilities are exchanged between peers during the TCP three way handshake. This patch adds support for this experimental option to TCP. References: [1] SMC-R Informational RFC: http://www.rfc-editor.org/info/rfc7609 [2] Shared Use of TCP Experimental Options RFC 6994: https://tools.ietf.org/rfc/rfc6994.txt [3] IANA ExID SMCR: http://www.iana.org/assignments/tcp-parameters/tcp-parameters.xhtml#tcp-exids Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- include/linux/tcp.h | 9 +++++-- include/net/inet_sock.h | 3 ++- include/net/tcp.h | 7 ++++++ net/ipv4/tcp.c | 6 +++++ net/ipv4/tcp_input.c | 35 +++++++++++++++++++++++++++ net/ipv4/tcp_minisocks.c | 19 +++++++++++++++ net/ipv4/tcp_output.c | 63 +++++++++++++++++++++++++++++++++++++++++++++--- 7 files changed, 136 insertions(+), 6 deletions(-) (limited to 'include/net') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 173a7c2f9636..8c431385b272 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -98,7 +98,8 @@ struct tcp_options_received { tstamp_ok : 1, /* TIMESTAMP seen on SYN packet */ dsack : 1, /* D-SACK is scheduled */ wscale_ok : 1, /* Wscale seen on SYN packet */ - sack_ok : 4, /* SACK seen on SYN packet */ + sack_ok : 3, /* SACK seen on SYN packet */ + smc_ok : 1, /* SMC seen on SYN packet */ snd_wscale : 4, /* Window scaling received from sender */ rcv_wscale : 4; /* Window scaling to send to receiver */ u8 num_sacks; /* Number of SACK blocks */ @@ -110,6 +111,9 @@ static inline void tcp_clear_options(struct tcp_options_received *rx_opt) { rx_opt->tstamp_ok = rx_opt->sack_ok = 0; rx_opt->wscale_ok = rx_opt->snd_wscale = 0; +#if IS_ENABLED(CONFIG_SMC) + rx_opt->smc_ok = 0; +#endif } /* This is the max number of SACKS that we'll generate and process. It's safe @@ -229,7 +233,8 @@ struct tcp_sock { syn_fastopen_ch:1, /* Active TFO re-enabling probe */ syn_data_acked:1,/* data in SYN is acked by SYN-ACK */ save_syn:1, /* Save headers of SYN packet */ - is_cwnd_limited:1;/* forward progress limited by snd_cwnd? */ + is_cwnd_limited:1,/* forward progress limited by snd_cwnd? */ + syn_smc:1; /* SYN includes SMC */ u32 tlp_high_seq; /* snd_nxt at the time of TLP retransmit. */ /* RTT measurement */ diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h index 425752f768d2..c49938d1481a 100644 --- a/include/net/inet_sock.h +++ b/include/net/inet_sock.h @@ -92,7 +92,8 @@ struct inet_request_sock { wscale_ok : 1, ecn_ok : 1, acked : 1, - no_srccheck: 1; + no_srccheck: 1, + smc_ok : 1; kmemcheck_bitfield_end(flags); u32 ir_mark; union { diff --git a/include/net/tcp.h b/include/net/tcp.h index 2392f74074e7..285bc82dea41 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -191,6 +191,7 @@ void tcp_time_wait(struct sock *sk, int state, int timeo); * experimental options. See draft-ietf-tcpm-experimental-options-00.txt */ #define TCPOPT_FASTOPEN_MAGIC 0xF989 +#define TCPOPT_SMC_MAGIC 0xE2D4C3D9 /* * TCP option lengths @@ -203,6 +204,7 @@ void tcp_time_wait(struct sock *sk, int state, int timeo); #define TCPOLEN_MD5SIG 18 #define TCPOLEN_FASTOPEN_BASE 2 #define TCPOLEN_EXP_FASTOPEN_BASE 4 +#define TCPOLEN_EXP_SMC_BASE 6 /* But this is what stacks really send out. */ #define TCPOLEN_TSTAMP_ALIGNED 12 @@ -213,6 +215,7 @@ void tcp_time_wait(struct sock *sk, int state, int timeo); #define TCPOLEN_SACK_PERBLOCK 8 #define TCPOLEN_MD5SIG_ALIGNED 20 #define TCPOLEN_MSS_ALIGNED 4 +#define TCPOLEN_EXP_SMC_BASE_ALIGNED 8 /* Flags in tp->nonagle */ #define TCP_NAGLE_OFF 1 /* Nagle's algo is disabled */ @@ -2108,4 +2111,8 @@ static inline bool tcp_bpf_ca_needs_ecn(struct sock *sk) { return (tcp_call_bpf(sk, BPF_SOCK_OPS_NEEDS_ECN) == 1); } + +#if IS_ENABLED(CONFIG_SMC) +extern struct static_key_false tcp_have_smc; +#endif #endif /* _TCP_H */ diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 8f36277e82e9..f6e1c00e300e 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -270,6 +270,7 @@ #include #include #include +#include #include #include @@ -302,6 +303,11 @@ EXPORT_SYMBOL(sysctl_tcp_wmem); atomic_long_t tcp_memory_allocated; /* Current allocated memory. */ EXPORT_SYMBOL(tcp_memory_allocated); +#if IS_ENABLED(CONFIG_SMC) +DEFINE_STATIC_KEY_FALSE(tcp_have_smc); +EXPORT_SYMBOL(tcp_have_smc); +#endif + /* * Current number of TCP sockets. */ diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 893286db4623..337f6011528a 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -76,6 +76,8 @@ #include #include #include +#include +#include int sysctl_tcp_fack __read_mostly; int sysctl_tcp_max_reordering __read_mostly = 300; @@ -3737,6 +3739,21 @@ static void tcp_parse_fastopen_option(int len, const unsigned char *cookie, foc->exp = exp_opt; } +static void smc_parse_options(const struct tcphdr *th, + struct tcp_options_received *opt_rx, + const unsigned char *ptr, + int opsize) +{ +#if IS_ENABLED(CONFIG_SMC) + if (static_branch_unlikely(&tcp_have_smc)) { + if (th->syn && !(opsize & 1) && + opsize >= TCPOLEN_EXP_SMC_BASE && + get_unaligned_be32(ptr) == TCPOPT_SMC_MAGIC) + opt_rx->smc_ok = 1; + } +#endif +} + /* Look for tcp options. Normally only called on SYN and SYNACK packets. * But, this can also be called on packets in the established flow when * the fast version below fails. @@ -3844,6 +3861,9 @@ void tcp_parse_options(const struct net *net, tcp_parse_fastopen_option(opsize - TCPOLEN_EXP_FASTOPEN_BASE, ptr + 2, th->syn, foc, true); + else + smc_parse_options(th, opt_rx, ptr, + opsize); break; } @@ -5598,6 +5618,16 @@ static bool tcp_rcv_fastopen_synack(struct sock *sk, struct sk_buff *synack, return false; } +static void smc_check_reset_syn(struct tcp_sock *tp) +{ +#if IS_ENABLED(CONFIG_SMC) + if (static_branch_unlikely(&tcp_have_smc)) { + if (tp->syn_smc && !tp->rx_opt.smc_ok) + tp->syn_smc = 0; + } +#endif +} + static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, const struct tcphdr *th) { @@ -5704,6 +5734,8 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, * is initialized. */ tp->copied_seq = tp->rcv_nxt; + smc_check_reset_syn(tp); + smp_mb(); tcp_finish_connect(sk, skb); @@ -6157,6 +6189,9 @@ static void tcp_openreq_init(struct request_sock *req, ireq->ir_rmt_port = tcp_hdr(skb)->source; ireq->ir_num = ntohs(tcp_hdr(skb)->dest); ireq->ir_mark = inet_request_mark(sk, skb); +#if IS_ENABLED(CONFIG_SMC) + ireq->smc_ok = rx_opt->smc_ok; +#endif } struct request_sock *inet_reqsk_alloc(const struct request_sock_ops *ops, diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index a952357054f4..056009f1c14f 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include #include @@ -416,6 +417,21 @@ void tcp_ca_openreq_child(struct sock *sk, const struct dst_entry *dst) } EXPORT_SYMBOL_GPL(tcp_ca_openreq_child); +static void smc_check_reset_syn_req(struct tcp_sock *oldtp, + struct request_sock *req, + struct tcp_sock *newtp) +{ +#if IS_ENABLED(CONFIG_SMC) + struct inet_request_sock *ireq; + + if (static_branch_unlikely(&tcp_have_smc)) { + ireq = inet_rsk(req); + if (oldtp->syn_smc && !ireq->smc_ok) + newtp->syn_smc = 0; + } +#endif +} + /* This is not only more efficient than what we used to do, it eliminates * a lot of code duplication between IPv4/IPv6 SYN recv processing. -DaveM * @@ -433,6 +449,9 @@ struct sock *tcp_create_openreq_child(const struct sock *sk, struct tcp_request_sock *treq = tcp_rsk(req); struct inet_connection_sock *newicsk = inet_csk(newsk); struct tcp_sock *newtp = tcp_sk(newsk); + struct tcp_sock *oldtp = tcp_sk(sk); + + smc_check_reset_syn_req(oldtp, req, newtp); /* Now setup tcp_sock */ newtp->pred_flags = 0; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 1f01f4c9c738..c8fc512e0bbb 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -41,6 +41,7 @@ #include #include #include +#include #include @@ -422,6 +423,22 @@ static inline bool tcp_urg_mode(const struct tcp_sock *tp) #define OPTION_MD5 (1 << 2) #define OPTION_WSCALE (1 << 3) #define OPTION_FAST_OPEN_COOKIE (1 << 8) +#define OPTION_SMC (1 << 9) + +static void smc_options_write(__be32 *ptr, u16 *options) +{ +#if IS_ENABLED(CONFIG_SMC) + if (static_branch_unlikely(&tcp_have_smc)) { + if (unlikely(OPTION_SMC & *options)) { + *ptr++ = htonl((TCPOPT_NOP << 24) | + (TCPOPT_NOP << 16) | + (TCPOPT_EXP << 8) | + (TCPOLEN_EXP_SMC_BASE)); + *ptr++ = htonl(TCPOPT_SMC_MAGIC); + } + } +#endif +} struct tcp_out_options { u16 options; /* bit field of OPTION_* */ @@ -540,6 +557,41 @@ static void tcp_options_write(__be32 *ptr, struct tcp_sock *tp, } ptr += (len + 3) >> 2; } + + smc_options_write(ptr, &options); +} + +static void smc_set_option(const struct tcp_sock *tp, + struct tcp_out_options *opts, + unsigned int *remaining) +{ +#if IS_ENABLED(CONFIG_SMC) + if (static_branch_unlikely(&tcp_have_smc)) { + if (tp->syn_smc) { + if (*remaining >= TCPOLEN_EXP_SMC_BASE_ALIGNED) { + opts->options |= OPTION_SMC; + *remaining -= TCPOLEN_EXP_SMC_BASE_ALIGNED; + } + } + } +#endif +} + +static void smc_set_option_cond(const struct tcp_sock *tp, + const struct inet_request_sock *ireq, + struct tcp_out_options *opts, + unsigned int *remaining) +{ +#if IS_ENABLED(CONFIG_SMC) + if (static_branch_unlikely(&tcp_have_smc)) { + if (tp->syn_smc && ireq->smc_ok) { + if (*remaining >= TCPOLEN_EXP_SMC_BASE_ALIGNED) { + opts->options |= OPTION_SMC; + *remaining -= TCPOLEN_EXP_SMC_BASE_ALIGNED; + } + } + } +#endif } /* Compute TCP options for SYN packets. This is not the final @@ -607,11 +659,14 @@ static unsigned int tcp_syn_options(struct sock *sk, struct sk_buff *skb, } } + smc_set_option(tp, opts, &remaining); + return MAX_TCP_OPTION_SPACE - remaining; } /* Set up TCP options for SYN-ACKs. */ -static unsigned int tcp_synack_options(struct request_sock *req, +static unsigned int tcp_synack_options(const struct sock *sk, + struct request_sock *req, unsigned int mss, struct sk_buff *skb, struct tcp_out_options *opts, const struct tcp_md5sig_key *md5, @@ -667,6 +722,8 @@ static unsigned int tcp_synack_options(struct request_sock *req, } } + smc_set_option_cond(tcp_sk(sk), ireq, opts, &remaining); + return MAX_TCP_OPTION_SPACE - remaining; } @@ -3195,8 +3252,8 @@ struct sk_buff *tcp_make_synack(const struct sock *sk, struct dst_entry *dst, md5 = tcp_rsk(req)->af_specific->req_md5_lookup(sk, req_to_sk(req)); #endif skb_set_hash(skb, tcp_rsk(req)->txhash, PKT_HASH_TYPE_L4); - tcp_header_size = tcp_synack_options(req, mss, skb, &opts, md5, foc) + - sizeof(*th); + tcp_header_size = tcp_synack_options(sk, req, mss, skb, &opts, md5, + foc) + sizeof(*th); skb_push(skb, tcp_header_size); skb_reset_transport_header(skb); -- cgit v1.2.3 From 32d18ab1d44166cbb1dcaf8b359183636734ddb1 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 24 Oct 2017 12:41:01 +0200 Subject: net: updating dst lastusage is an unlikely event. Since commit 0da4af00b2ed ("ipv6: only update __use and lastusetime once per jiffy at most"), updating the dst lastuse field is an unlikely action: it happens at most once per jiffy, out of potentially millions of calls per second. Mark explicitly the code as such, and let the compiler generate better code. Note: gcc 7.2 and several older versions do actually generate different - better - code when the unlikely() hint is in place, avoid jump in the fast path and keeping better code locality. Signed-off-by: Paolo Abeni Signed-off-by: David S. Miller --- include/net/dst.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/dst.h b/include/net/dst.h index 5047e8053d6c..2f53ecc2c296 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -257,7 +257,7 @@ static inline void dst_hold(struct dst_entry *dst) static inline void dst_use_noref(struct dst_entry *dst, unsigned long time) { - if (time != dst->lastuse) { + if (unlikely(time != dst->lastuse)) { dst->__use++; dst->lastuse = time; } -- cgit v1.2.3 From 2ae21cf527da0e5cf9d7ee14bd5b0909bb9d1a75 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Oct 2017 21:54:56 -0700 Subject: tcp: Namespace-ify sysctl_tcp_early_retrans Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 18 +++++++++--------- net/ipv4/tcp_input.c | 1 - net/ipv4/tcp_ipv4.c | 1 + net/ipv4/tcp_output.c | 4 +++- 6 files changed, 14 insertions(+), 12 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 2c4222a5d102..a7f39e3ea666 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -128,6 +128,7 @@ struct netns_ipv4 { int sysctl_tcp_sack; int sysctl_tcp_window_scaling; int sysctl_tcp_timestamps; + int sysctl_tcp_early_retrans; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 285bc82dea41..a12b71d4118b 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -265,7 +265,6 @@ extern int sysctl_tcp_workaround_signed_windows; extern int sysctl_tcp_slow_start_after_idle; extern int sysctl_tcp_thin_linear_timeouts; extern int sysctl_tcp_thin_dupack; -extern int sysctl_tcp_early_retrans; extern int sysctl_tcp_recovery; #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 81d218346cf7..f0f650f020af 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -634,15 +634,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "tcp_early_retrans", - .data = &sysctl_tcp_early_retrans, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &zero, - .extra2 = &four, - }, { .procname = "tcp_min_tso_segs", .data = &sysctl_tcp_min_tso_segs, @@ -1145,6 +1136,15 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_early_retrans", + .data = &init_net.ipv4.sysctl_tcp_early_retrans, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &four, + }, { } }; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 337f6011528a..7656b1e6d504 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -95,7 +95,6 @@ int sysctl_tcp_max_orphans __read_mostly = NR_FILE; int sysctl_tcp_frto __read_mostly = 2; int sysctl_tcp_min_rtt_wlen __read_mostly = 300; int sysctl_tcp_moderate_rcvbuf __read_mostly = 1; -int sysctl_tcp_early_retrans __read_mostly = 3; int sysctl_tcp_invalid_ratelimit __read_mostly = HZ/2; #define FLAG_DATA 0x01 /* Incoming frame contained data. */ diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 23a8100af5ad..7ab313f6768e 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2484,6 +2484,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_sack = 1; net->ipv4.sysctl_tcp_window_scaling = 1; net->ipv4.sysctl_tcp_timestamps = 1; + net->ipv4.sysctl_tcp_early_retrans = 3; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index c8fc512e0bbb..21713836d46a 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2435,6 +2435,7 @@ bool tcp_schedule_loss_probe(struct sock *sk) struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); u32 timeout, rto_delta_us; + int early_retrans; /* Don't do any loss probe on a Fast Open connection before 3WHS * finishes. @@ -2442,10 +2443,11 @@ bool tcp_schedule_loss_probe(struct sock *sk) if (tp->fastopen_rsk) return false; + early_retrans = sock_net(sk)->ipv4.sysctl_tcp_early_retrans; /* Schedule a loss probe in 2*RTT for SACK capable connections * in Open state, that are either limited by cwnd or application. */ - if ((sysctl_tcp_early_retrans != 3 && sysctl_tcp_early_retrans != 4) || + if ((early_retrans != 3 && early_retrans != 4) || !tp->packets_out || !tcp_is_sack(tp) || icsk->icsk_ca_state != TCP_CA_Open) return false; -- cgit v1.2.3 From e20223f1962831d1b1c416d59d259879d0639d68 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Oct 2017 21:54:57 -0700 Subject: tcp: Namespace-ify sysctl_tcp_recovery Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 2 +- net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_input.c | 2 +- net/ipv4/tcp_ipv4.c | 1 + net/ipv4/tcp_recovery.c | 2 -- 6 files changed, 11 insertions(+), 11 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index a7f39e3ea666..d6ed718075d4 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -129,6 +129,7 @@ struct netns_ipv4 { int sysctl_tcp_window_scaling; int sysctl_tcp_timestamps; int sysctl_tcp_early_retrans; + int sysctl_tcp_recovery; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index a12b71d4118b..c7f51534fc44 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -265,7 +265,7 @@ extern int sysctl_tcp_workaround_signed_windows; extern int sysctl_tcp_slow_start_after_idle; extern int sysctl_tcp_thin_linear_timeouts; extern int sysctl_tcp_thin_dupack; -extern int sysctl_tcp_recovery; + #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ extern int sysctl_tcp_limit_output_bytes; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index f0f650f020af..78019adcae87 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -449,13 +449,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "tcp_recovery", - .data = &sysctl_tcp_recovery, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec, - }, { .procname = "tcp_max_reordering", .data = &sysctl_tcp_max_reordering, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .extra1 = &zero, .extra2 = &four, }, + { + .procname = "tcp_recovery", + .data = &init_net.ipv4.sysctl_tcp_recovery, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, { } }; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 7656b1e6d504..5b2272dbf6a9 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -2788,7 +2788,7 @@ static void tcp_rack_identify_loss(struct sock *sk, int *ack_flag) struct tcp_sock *tp = tcp_sk(sk); /* Use RACK to detect loss */ - if (sysctl_tcp_recovery & TCP_RACK_LOSS_DETECTION) { + if (sock_net(sk)->ipv4.sysctl_tcp_recovery & TCP_RACK_LOSS_DETECTION) { u32 prior_retrans = tp->retrans_out; tcp_rack_mark_lost(sk); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 7ab313f6768e..517ff1948a71 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2485,6 +2485,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_window_scaling = 1; net->ipv4.sysctl_tcp_timestamps = 1; net->ipv4.sysctl_tcp_early_retrans = 3; + net->ipv4.sysctl_tcp_recovery = TCP_RACK_LOSS_DETECTION; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); diff --git a/net/ipv4/tcp_recovery.c b/net/ipv4/tcp_recovery.c index cda6074a429a..d3603a9e24ea 100644 --- a/net/ipv4/tcp_recovery.c +++ b/net/ipv4/tcp_recovery.c @@ -1,8 +1,6 @@ #include #include -int sysctl_tcp_recovery __read_mostly = TCP_RACK_LOSS_DETECTION; - static void tcp_rack_mark_skb_lost(struct sock *sk, struct sk_buff *skb) { struct tcp_sock *tp = tcp_sk(sk); -- cgit v1.2.3 From 2c04ac8ae0b61e0780a30b7069a11bb202b21f26 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Oct 2017 21:54:58 -0700 Subject: tcp: Namespace-ify sysctl_tcp_thin_linear_timeouts Note that sysctl_tcp_thin_dupack was not used, I deleted it. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 2 -- net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_timer.c | 4 +--- 4 files changed, 9 insertions(+), 12 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index d6ed718075d4..2a9f37b39c45 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -130,6 +130,7 @@ struct netns_ipv4 { int sysctl_tcp_timestamps; int sysctl_tcp_early_retrans; int sysctl_tcp_recovery; + int sysctl_tcp_thin_linear_timeouts; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index c7f51534fc44..063a7a48b7fe 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -263,8 +263,6 @@ extern int sysctl_tcp_moderate_rcvbuf; extern int sysctl_tcp_tso_win_divisor; extern int sysctl_tcp_workaround_signed_windows; extern int sysctl_tcp_slow_start_after_idle; -extern int sysctl_tcp_thin_linear_timeouts; -extern int sysctl_tcp_thin_dupack; #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 78019adcae87..12003214f4d8 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -620,13 +620,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_allowed_congestion_control, }, - { - .procname = "tcp_thin_linear_timeouts", - .data = &sysctl_tcp_thin_linear_timeouts, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, { .procname = "tcp_min_tso_segs", .data = &sysctl_tcp_min_tso_segs, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, + { + .procname = "tcp_thin_linear_timeouts", + .data = &init_net.ipv4.sysctl_tcp_thin_linear_timeouts, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 804a8d34ce86..035a1ef1f2d8 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -22,8 +22,6 @@ #include #include -int sysctl_tcp_thin_linear_timeouts __read_mostly; - /** * tcp_write_err() - close socket and save error info * @sk: The socket the error has appeared on. @@ -522,7 +520,7 @@ out_reset_timer: * linear-timeout retransmissions into a black hole */ if (sk->sk_state == TCP_ESTABLISHED && - (tp->thin_lto || sysctl_tcp_thin_linear_timeouts) && + (tp->thin_lto || net->ipv4.sysctl_tcp_thin_linear_timeouts) && tcp_stream_is_thin(tp) && icsk->icsk_retransmits <= TCP_THIN_LINEAR_RETRIES) { icsk->icsk_backoff = 0; -- cgit v1.2.3 From b510f0d23a47c3d1f074fe583e7867dc4918fe02 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Oct 2017 21:54:59 -0700 Subject: tcp: Namespace-ify sysctl_tcp_slow_start_after_idle Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 3 +-- net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_ipv4.c | 1 + net/ipv4/tcp_output.c | 5 +---- 5 files changed, 11 insertions(+), 13 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 2a9f37b39c45..8662692686b3 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -131,6 +131,7 @@ struct netns_ipv4 { int sysctl_tcp_early_retrans; int sysctl_tcp_recovery; int sysctl_tcp_thin_linear_timeouts; + int sysctl_tcp_slow_start_after_idle; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 063a7a48b7fe..cc2ab522eb5c 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -262,7 +262,6 @@ extern int sysctl_tcp_nometrics_save; extern int sysctl_tcp_moderate_rcvbuf; extern int sysctl_tcp_tso_win_divisor; extern int sysctl_tcp_workaround_signed_windows; -extern int sysctl_tcp_slow_start_after_idle; #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ @@ -1308,7 +1307,7 @@ static inline void tcp_slow_start_after_idle_check(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); s32 delta; - if (!sysctl_tcp_slow_start_after_idle || tp->packets_out || + if (!sock_net(sk)->ipv4.sysctl_tcp_slow_start_after_idle || tp->packets_out || ca_ops->cong_control) return; delta = tcp_jiffies32 - tp->lsndtime; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 12003214f4d8..40d69af8b363 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -571,13 +571,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "tcp_slow_start_after_idle", - .data = &sysctl_tcp_slow_start_after_idle, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, #ifdef CONFIG_NETLABEL { .procname = "cipso_cache_enable", @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_slow_start_after_idle", + .data = &init_net.ipv4.sysctl_tcp_slow_start_after_idle, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 517ff1948a71..cea63a4b5965 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2486,6 +2486,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_timestamps = 1; net->ipv4.sysctl_tcp_early_retrans = 3; net->ipv4.sysctl_tcp_recovery = TCP_RACK_LOSS_DETECTION; + net->ipv4.sysctl_tcp_slow_start_after_idle = 1; /* By default, RFC2861 behavior. */ net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 21713836d46a..bdc288a06f94 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -62,9 +62,6 @@ int sysctl_tcp_limit_output_bytes __read_mostly = 262144; */ int sysctl_tcp_tso_win_divisor __read_mostly = 3; -/* By default, RFC2861 behavior. */ -int sysctl_tcp_slow_start_after_idle __read_mostly = 1; - static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, int push_one, gfp_t gfp); @@ -1690,7 +1687,7 @@ static void tcp_cwnd_validate(struct sock *sk, bool is_cwnd_limited) if (tp->packets_out > tp->snd_cwnd_used) tp->snd_cwnd_used = tp->packets_out; - if (sysctl_tcp_slow_start_after_idle && + if (sock_net(sk)->ipv4.sysctl_tcp_slow_start_after_idle && (s32)(tcp_jiffies32 - tp->snd_cwnd_stamp) >= inet_csk(sk)->icsk_rto && !ca_ops->cong_control) tcp_cwnd_application_limited(sk); -- cgit v1.2.3 From e0a1e5b519236dc1662ff25e42560dd1be9e3776 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Oct 2017 21:55:00 -0700 Subject: tcp: Namespace-ify sysctl_tcp_retrans_collapse Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_ipv4.c | 2 +- net/ipv4/tcp_output.c | 5 +---- 5 files changed, 10 insertions(+), 13 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 8662692686b3..b28c172b10e4 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -132,6 +132,7 @@ struct netns_ipv4 { int sysctl_tcp_recovery; int sysctl_tcp_thin_linear_timeouts; int sysctl_tcp_slow_start_after_idle; + int sysctl_tcp_retrans_collapse; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index cc2ab522eb5c..33cc86355b8f 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -243,7 +243,6 @@ void tcp_time_wait(struct sock *sk, int state, int timeo); /* sysctl variables for tcp */ -extern int sysctl_tcp_retrans_collapse; extern int sysctl_tcp_stdurg; extern int sysctl_tcp_rfc1337; extern int sysctl_tcp_abort_on_overflow; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 40d69af8b363..533b92ad39dd 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -386,13 +386,6 @@ static int proc_tcp_available_ulp(struct ctl_table *ctl, } static struct ctl_table ipv4_table[] = { - { - .procname = "tcp_retrans_collapse", - .data = &sysctl_tcp_retrans_collapse, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, { .procname = "tcp_max_orphans", .data = &sysctl_tcp_max_orphans, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_retrans_collapse", + .data = &init_net.ipv4.sysctl_tcp_retrans_collapse, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index cea63a4b5965..2bc6ba2059d3 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2487,7 +2487,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_early_retrans = 3; net->ipv4.sysctl_tcp_recovery = TCP_RACK_LOSS_DETECTION; net->ipv4.sysctl_tcp_slow_start_after_idle = 1; /* By default, RFC2861 behavior. */ - + net->ipv4.sysctl_tcp_retrans_collapse = 1; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); net->ipv4.sysctl_tcp_fastopen_blackhole_timeout = 60 * 60; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index bdc288a06f94..55a0aa4b96df 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -45,9 +45,6 @@ #include -/* People can turn this off for buggy TCP's found in printers etc. */ -int sysctl_tcp_retrans_collapse __read_mostly = 1; - /* People can turn this on to work with those rare, broken TCPs that * interpret the window field as a signed quantity. */ @@ -2804,7 +2801,7 @@ static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *to, struct sk_buff *skb = to, *tmp; bool first = true; - if (!sysctl_tcp_retrans_collapse) + if (!sock_net(sk)->ipv4.sysctl_tcp_retrans_collapse) return; if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_SYN) return; -- cgit v1.2.3 From 3f4c7c6f6a9053493ce7dd8a0f17ed8eafc53893 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Oct 2017 21:55:01 -0700 Subject: tcp: Namespace-ify sysctl_tcp_stdurg Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_input.c | 3 +-- 4 files changed, 9 insertions(+), 10 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index b28c172b10e4..ffa2cf3dc747 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -133,6 +133,7 @@ struct netns_ipv4 { int sysctl_tcp_thin_linear_timeouts; int sysctl_tcp_slow_start_after_idle; int sysctl_tcp_retrans_collapse; + int sysctl_tcp_stdurg; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 33cc86355b8f..cf3fac7008d7 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -243,7 +243,6 @@ void tcp_time_wait(struct sock *sk, int state, int timeo); /* sysctl variables for tcp */ -extern int sysctl_tcp_stdurg; extern int sysctl_tcp_rfc1337; extern int sysctl_tcp_abort_on_overflow; extern int sysctl_tcp_max_orphans; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 533b92ad39dd..a34bb75815c1 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -400,13 +400,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "tcp_stdurg", - .data = &sysctl_tcp_stdurg, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, { .procname = "tcp_rfc1337", .data = &sysctl_tcp_rfc1337, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_stdurg", + .data = &init_net.ipv4.sysctl_tcp_stdurg, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 5b2272dbf6a9..14b06963c102 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -89,7 +89,6 @@ EXPORT_SYMBOL(sysctl_tcp_adv_win_scale); /* rfc5961 challenge ack rate limiting */ int sysctl_tcp_challenge_ack_limit = 1000; -int sysctl_tcp_stdurg __read_mostly; int sysctl_tcp_rfc1337 __read_mostly; int sysctl_tcp_max_orphans __read_mostly = NR_FILE; int sysctl_tcp_frto __read_mostly = 2; @@ -5123,7 +5122,7 @@ static void tcp_check_urg(struct sock *sk, const struct tcphdr *th) struct tcp_sock *tp = tcp_sk(sk); u32 ptr = ntohs(th->urg_ptr); - if (ptr && !sysctl_tcp_stdurg) + if (ptr && !sock_net(sk)->ipv4.sysctl_tcp_stdurg) ptr--; ptr += ntohl(th->seq); -- cgit v1.2.3 From 625357aa175c688d219da43c8cfaa2e1629e0e1a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Oct 2017 21:55:02 -0700 Subject: tcp: Namespace-ify sysctl_tcp_rfc1337 Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_input.c | 1 - net/ipv4/tcp_minisocks.c | 2 +- 5 files changed, 9 insertions(+), 10 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index ffa2cf3dc747..968edce38eb5 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -134,6 +134,7 @@ struct netns_ipv4 { int sysctl_tcp_slow_start_after_idle; int sysctl_tcp_retrans_collapse; int sysctl_tcp_stdurg; + int sysctl_tcp_rfc1337; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index cf3fac7008d7..2aea2b3373b3 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -243,7 +243,6 @@ void tcp_time_wait(struct sock *sk, int state, int timeo); /* sysctl variables for tcp */ -extern int sysctl_tcp_rfc1337; extern int sysctl_tcp_abort_on_overflow; extern int sysctl_tcp_max_orphans; extern int sysctl_tcp_fack; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index a34bb75815c1..832e554235df 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -400,13 +400,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "tcp_rfc1337", - .data = &sysctl_tcp_rfc1337, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, { .procname = "inet_peer_threshold", .data = &inet_peer_threshold, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_rfc1337", + .data = &init_net.ipv4.sysctl_tcp_rfc1337, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 14b06963c102..64fde81b0eb7 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -89,7 +89,6 @@ EXPORT_SYMBOL(sysctl_tcp_adv_win_scale); /* rfc5961 challenge ack rate limiting */ int sysctl_tcp_challenge_ack_limit = 1000; -int sysctl_tcp_rfc1337 __read_mostly; int sysctl_tcp_max_orphans __read_mostly = NR_FILE; int sysctl_tcp_frto __read_mostly = 2; int sysctl_tcp_min_rtt_wlen __read_mostly = 300; diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 056009f1c14f..11836667763c 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -181,7 +181,7 @@ tcp_timewait_state_process(struct inet_timewait_sock *tw, struct sk_buff *skb, * Oh well... nobody has a sufficient solution to this * protocol bug yet. */ - if (sysctl_tcp_rfc1337 == 0) { + if (twsk_net(tw)->ipv4.sysctl_tcp_rfc1337 == 0) { kill: inet_twsk_deschedule_put(tw); return TCP_TW_SUCCESS; -- cgit v1.2.3 From 65c9410cf55ecf32da1b720f563365d565d6289a Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Oct 2017 21:55:03 -0700 Subject: tcp: Namespace-ify sysctl_tcp_abort_on_overflow Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_minisocks.c | 4 +--- 4 files changed, 9 insertions(+), 11 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 968edce38eb5..3875fdf6b186 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -135,6 +135,7 @@ struct netns_ipv4 { int sysctl_tcp_retrans_collapse; int sysctl_tcp_stdurg; int sysctl_tcp_rfc1337; + int sysctl_tcp_abort_on_overflow; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 2aea2b3373b3..7331281a2292 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -243,7 +243,6 @@ void tcp_time_wait(struct sock *sk, int state, int timeo); /* sysctl variables for tcp */ -extern int sysctl_tcp_abort_on_overflow; extern int sysctl_tcp_max_orphans; extern int sysctl_tcp_fack; extern int sysctl_tcp_reordering; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 832e554235df..ffd1fd769bba 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -393,13 +393,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "tcp_abort_on_overflow", - .data = &sysctl_tcp_abort_on_overflow, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, { .procname = "inet_peer_threshold", .data = &inet_peer_threshold, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_abort_on_overflow", + .data = &init_net.ipv4.sysctl_tcp_abort_on_overflow, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 11836667763c..3674d63170b2 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -29,8 +29,6 @@ #include #include -int sysctl_tcp_abort_on_overflow __read_mostly; - static bool tcp_in_window(u32 seq, u32 end_seq, u32 s_win, u32 e_win) { if (seq == s_win) @@ -783,7 +781,7 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb, return inet_csk_complete_hashdance(sk, child, req, own_req); listen_overflow: - if (!sysctl_tcp_abort_on_overflow) { + if (!sock_net(sk)->ipv4.sysctl_tcp_abort_on_overflow) { inet_rsk(req)->acked = 1; return NULL; } -- cgit v1.2.3 From 0bc65a28ae2aeb14aab7f4a930e0d8cf4cad9dc4 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Oct 2017 21:55:04 -0700 Subject: tcp: Namespace-ify sysctl_tcp_fack Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp.c | 2 +- net/ipv4/tcp_input.c | 3 +-- net/ipv4/tcp_minisocks.c | 2 +- 6 files changed, 11 insertions(+), 12 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 3875fdf6b186..f0e792beeea9 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -136,6 +136,7 @@ struct netns_ipv4 { int sysctl_tcp_stdurg; int sysctl_tcp_rfc1337; int sysctl_tcp_abort_on_overflow; + int sysctl_tcp_fack; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 7331281a2292..e7b15e9f6e28 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -244,7 +244,6 @@ void tcp_time_wait(struct sock *sk, int state, int timeo); /* sysctl variables for tcp */ extern int sysctl_tcp_max_orphans; -extern int sysctl_tcp_fack; extern int sysctl_tcp_reordering; extern int sysctl_tcp_max_reordering; extern int sysctl_tcp_dsack; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index ffd1fd769bba..1f23be13ce7b 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -414,13 +414,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { - .procname = "tcp_fack", - .data = &sysctl_tcp_fack, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, { .procname = "tcp_max_reordering", .data = &sysctl_tcp_max_reordering, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_fack", + .data = &init_net.ipv4.sysctl_tcp_fack, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index f6e1c00e300e..c7c983f0f817 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2517,7 +2517,7 @@ static int tcp_repair_options_est(struct sock *sk, return -EINVAL; tp->rx_opt.sack_ok |= TCP_SACK_SEEN; - if (sysctl_tcp_fack) + if (sock_net(sk)->ipv4.sysctl_tcp_fack) tcp_enable_fack(tp); break; case TCPOPT_TIMESTAMP: diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 64fde81b0eb7..c5b94460793f 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -79,7 +79,6 @@ #include #include -int sysctl_tcp_fack __read_mostly; int sysctl_tcp_max_reordering __read_mostly = 300; int sysctl_tcp_dsack __read_mostly = 1; int sysctl_tcp_app_win __read_mostly = 31; @@ -5720,7 +5719,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, tp->tcp_header_len = sizeof(struct tcphdr); } - if (tcp_is_sack(tp) && sysctl_tcp_fack) + if (tcp_is_sack(tp) && sock_net(sk)->ipv4.sysctl_tcp_fack) tcp_enable_fack(tp); tcp_sync_mss(sk, icsk->icsk_pmtu_cookie); diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 3674d63170b2..3270ab8416ce 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -510,7 +510,7 @@ struct sock *tcp_create_openreq_child(const struct sock *sk, newtp->rx_opt.tstamp_ok = ireq->tstamp_ok; if ((newtp->rx_opt.sack_ok = ireq->sack_ok) != 0) { - if (sysctl_tcp_fack) + if (sock_net(sk)->ipv4.sysctl_tcp_fack) tcp_enable_fack(newtp); } newtp->window_clamp = req->rsk_window_clamp; -- cgit v1.2.3 From 773d4bb96ceca6829ae9928f6b002b93e2e62cdc Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Oct 2017 21:55:05 -0700 Subject: tcp: remove stale sysctl_tcp_reordering This extern is no longer used. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/tcp.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/net') diff --git a/include/net/tcp.h b/include/net/tcp.h index e7b15e9f6e28..fc134ba74c7d 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -244,7 +244,6 @@ void tcp_time_wait(struct sock *sk, int state, int timeo); /* sysctl variables for tcp */ extern int sysctl_tcp_max_orphans; -extern int sysctl_tcp_reordering; extern int sysctl_tcp_max_reordering; extern int sysctl_tcp_dsack; extern long sysctl_tcp_mem[3]; -- cgit v1.2.3 From c6e218035913e14952b04ceecf1a543205106fdb Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Oct 2017 21:55:06 -0700 Subject: tcp: Namespace-ify sysctl_tcp_max_reordering Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_input.c | 3 +-- net/ipv4/tcp_ipv4.c | 2 ++ 5 files changed, 11 insertions(+), 10 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index f0e792beeea9..3f6844665a2f 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -137,6 +137,7 @@ struct netns_ipv4 { int sysctl_tcp_rfc1337; int sysctl_tcp_abort_on_overflow; int sysctl_tcp_fack; + int sysctl_tcp_max_reordering; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index fc134ba74c7d..8cd286226a1e 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -244,7 +244,6 @@ void tcp_time_wait(struct sock *sk, int state, int timeo); /* sysctl variables for tcp */ extern int sysctl_tcp_max_orphans; -extern int sysctl_tcp_max_reordering; extern int sysctl_tcp_dsack; extern long sysctl_tcp_mem[3]; extern int sysctl_tcp_wmem[3]; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 1f23be13ce7b..18cd228a2069 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -414,13 +414,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { - .procname = "tcp_max_reordering", - .data = &sysctl_tcp_max_reordering, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, { .procname = "tcp_dsack", .data = &sysctl_tcp_dsack, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_max_reordering", + .data = &init_net.ipv4.sysctl_tcp_max_reordering, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index c5b94460793f..c118657f06ee 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -79,7 +79,6 @@ #include #include -int sysctl_tcp_max_reordering __read_mostly = 300; int sysctl_tcp_dsack __read_mostly = 1; int sysctl_tcp_app_win __read_mostly = 31; int sysctl_tcp_adv_win_scale __read_mostly = 1; @@ -889,7 +888,7 @@ static void tcp_update_reordering(struct sock *sk, const int metric, return; if (metric > tp->reordering) { - tp->reordering = min(sysctl_tcp_max_reordering, metric); + tp->reordering = min(sock_net(sk)->ipv4.sysctl_tcp_max_reordering, metric); #if FASTRETRANS_DEBUG > 1 pr_debug("Disorder%d %d %u f%u s%u rr%d\n", diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 2bc6ba2059d3..c379a242abb3 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2488,6 +2488,8 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_recovery = TCP_RACK_LOSS_DETECTION; net->ipv4.sysctl_tcp_slow_start_after_idle = 1; /* By default, RFC2861 behavior. */ net->ipv4.sysctl_tcp_retrans_collapse = 1; + net->ipv4.sysctl_tcp_max_reordering = 300; + net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); net->ipv4.sysctl_tcp_fastopen_blackhole_timeout = 60 * 60; -- cgit v1.2.3 From 6496f6bde0c323fba5e8c5b5cbf3a7bf28dad7ed Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Oct 2017 21:55:07 -0700 Subject: tcp: Namespace-ify sysctl_tcp_dsack Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_input.c | 5 ++--- net/ipv4/tcp_ipv4.c | 1 + 5 files changed, 11 insertions(+), 11 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 3f6844665a2f..956957a77db9 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -138,6 +138,7 @@ struct netns_ipv4 { int sysctl_tcp_abort_on_overflow; int sysctl_tcp_fack; int sysctl_tcp_max_reordering; + int sysctl_tcp_dsack; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 8cd286226a1e..8b2ae3e8d79f 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -244,7 +244,6 @@ void tcp_time_wait(struct sock *sk, int state, int timeo); /* sysctl variables for tcp */ extern int sysctl_tcp_max_orphans; -extern int sysctl_tcp_dsack; extern long sysctl_tcp_mem[3]; extern int sysctl_tcp_wmem[3]; extern int sysctl_tcp_rmem[3]; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 18cd228a2069..7652a9c2a65d 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -414,13 +414,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { - .procname = "tcp_dsack", - .data = &sysctl_tcp_dsack, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, { .procname = "tcp_mem", .maxlen = sizeof(sysctl_tcp_mem), @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_dsack", + .data = &init_net.ipv4.sysctl_tcp_dsack, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index c118657f06ee..fd77037ac800 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -79,7 +79,6 @@ #include #include -int sysctl_tcp_dsack __read_mostly = 1; int sysctl_tcp_app_win __read_mostly = 31; int sysctl_tcp_adv_win_scale __read_mostly = 1; EXPORT_SYMBOL(sysctl_tcp_adv_win_scale); @@ -4150,7 +4149,7 @@ static void tcp_dsack_set(struct sock *sk, u32 seq, u32 end_seq) { struct tcp_sock *tp = tcp_sk(sk); - if (tcp_is_sack(tp) && sysctl_tcp_dsack) { + if (tcp_is_sack(tp) && sock_net(sk)->ipv4.sysctl_tcp_dsack) { int mib_idx; if (before(seq, tp->rcv_nxt)) @@ -4185,7 +4184,7 @@ static void tcp_send_dupack(struct sock *sk, const struct sk_buff *skb) NET_INC_STATS(sock_net(sk), LINUX_MIB_DELAYEDACKLOST); tcp_enter_quickack_mode(sk); - if (tcp_is_sack(tp) && sysctl_tcp_dsack) { + if (tcp_is_sack(tp) && sock_net(sk)->ipv4.sysctl_tcp_dsack) { u32 end_seq = TCP_SKB_CB(skb)->end_seq; if (after(TCP_SKB_CB(skb)->end_seq, tp->rcv_nxt)) diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index c379a242abb3..d9d4d191e8f3 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2489,6 +2489,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_slow_start_after_idle = 1; /* By default, RFC2861 behavior. */ net->ipv4.sysctl_tcp_retrans_collapse = 1; net->ipv4.sysctl_tcp_max_reordering = 300; + net->ipv4.sysctl_tcp_dsack = 1; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); -- cgit v1.2.3 From 0c12654ac6d9004b9538b2a969b2b59e9a5ed831 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Oct 2017 21:55:08 -0700 Subject: tcp: Namespace-ify sysctl_tcp_app_win Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_input.c | 8 ++++---- net/ipv4/tcp_ipv4.c | 1 + 5 files changed, 13 insertions(+), 12 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 956957a77db9..63f91d52cbc0 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -139,6 +139,7 @@ struct netns_ipv4 { int sysctl_tcp_fack; int sysctl_tcp_max_reordering; int sysctl_tcp_dsack; + int sysctl_tcp_app_win; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 8b2ae3e8d79f..7aa3d65062a1 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -247,7 +247,6 @@ extern int sysctl_tcp_max_orphans; extern long sysctl_tcp_mem[3]; extern int sysctl_tcp_wmem[3]; extern int sysctl_tcp_rmem[3]; -extern int sysctl_tcp_app_win; extern int sysctl_tcp_adv_win_scale; extern int sysctl_tcp_frto; extern int sysctl_tcp_nometrics_save; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 7652a9c2a65d..e057788834a9 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -437,13 +437,6 @@ static struct ctl_table ipv4_table[] = { .proc_handler = proc_dointvec_minmax, .extra1 = &one, }, - { - .procname = "tcp_app_win", - .data = &sysctl_tcp_app_win, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, { .procname = "tcp_adv_win_scale", .data = &sysctl_tcp_adv_win_scale, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_app_win", + .data = &init_net.ipv4.sysctl_tcp_app_win, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index fd77037ac800..6af4b58ac6d5 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -79,7 +79,6 @@ #include #include -int sysctl_tcp_app_win __read_mostly = 31; int sysctl_tcp_adv_win_scale __read_mostly = 1; EXPORT_SYMBOL(sysctl_tcp_adv_win_scale); @@ -428,6 +427,7 @@ static void tcp_fixup_rcvbuf(struct sock *sk) */ void tcp_init_buffer_space(struct sock *sk) { + int tcp_app_win = sock_net(sk)->ipv4.sysctl_tcp_app_win; struct tcp_sock *tp = tcp_sk(sk); int maxwin; @@ -446,14 +446,14 @@ void tcp_init_buffer_space(struct sock *sk) if (tp->window_clamp >= maxwin) { tp->window_clamp = maxwin; - if (sysctl_tcp_app_win && maxwin > 4 * tp->advmss) + if (tcp_app_win && maxwin > 4 * tp->advmss) tp->window_clamp = max(maxwin - - (maxwin >> sysctl_tcp_app_win), + (maxwin >> tcp_app_win), 4 * tp->advmss); } /* Force reservation of one segment. */ - if (sysctl_tcp_app_win && + if (tcp_app_win && tp->window_clamp > 2 * tp->advmss && tp->window_clamp + tp->advmss > maxwin) tp->window_clamp = max(2 * tp->advmss, maxwin - tp->advmss); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index d9d4d191e8f3..189664ebd28e 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2490,6 +2490,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_retrans_collapse = 1; net->ipv4.sysctl_tcp_max_reordering = 300; net->ipv4.sysctl_tcp_dsack = 1; + net->ipv4.sysctl_tcp_app_win = 31; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); -- cgit v1.2.3 From 94f0893e0c27219f4a726932618505aab6795973 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Oct 2017 21:55:09 -0700 Subject: tcp: Namespace-ify sysctl_tcp_adv_win_scale Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 9 ++++----- net/ipv4/sysctl_net_ipv4.c | 18 +++++++++--------- net/ipv4/tcp_input.c | 13 +++++-------- net/ipv4/tcp_ipv4.c | 1 + 5 files changed, 20 insertions(+), 22 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 63f91d52cbc0..9dbb07d4eff4 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -140,6 +140,7 @@ struct netns_ipv4 { int sysctl_tcp_max_reordering; int sysctl_tcp_dsack; int sysctl_tcp_app_win; + int sysctl_tcp_adv_win_scale; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 7aa3d65062a1..0dc27cd24899 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -247,7 +247,6 @@ extern int sysctl_tcp_max_orphans; extern long sysctl_tcp_mem[3]; extern int sysctl_tcp_wmem[3]; extern int sysctl_tcp_rmem[3]; -extern int sysctl_tcp_adv_win_scale; extern int sysctl_tcp_frto; extern int sysctl_tcp_nometrics_save; extern int sysctl_tcp_moderate_rcvbuf; @@ -1311,9 +1310,9 @@ void tcp_select_initial_window(int __space, __u32 mss, __u32 *rcv_wnd, __u32 *window_clamp, int wscale_ok, __u8 *rcv_wscale, __u32 init_rcv_wnd); -static inline int tcp_win_from_space(int space) +static inline int tcp_win_from_space(const struct sock *sk, int space) { - int tcp_adv_win_scale = sysctl_tcp_adv_win_scale; + int tcp_adv_win_scale = sock_net(sk)->ipv4.sysctl_tcp_adv_win_scale; return tcp_adv_win_scale <= 0 ? (space>>(-tcp_adv_win_scale)) : @@ -1323,13 +1322,13 @@ static inline int tcp_win_from_space(int space) /* Note: caller must be prepared to deal with negative returns */ static inline int tcp_space(const struct sock *sk) { - return tcp_win_from_space(sk->sk_rcvbuf - + return tcp_win_from_space(sk, sk->sk_rcvbuf - atomic_read(&sk->sk_rmem_alloc)); } static inline int tcp_full_space(const struct sock *sk) { - return tcp_win_from_space(sk->sk_rcvbuf); + return tcp_win_from_space(sk, sk->sk_rcvbuf); } extern void tcp_openreq_init_rwin(struct request_sock *req, diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index e057788834a9..a95123e1e7da 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -437,15 +437,6 @@ static struct ctl_table ipv4_table[] = { .proc_handler = proc_dointvec_minmax, .extra1 = &one, }, - { - .procname = "tcp_adv_win_scale", - .data = &sysctl_tcp_adv_win_scale, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &tcp_adv_win_scale_min, - .extra2 = &tcp_adv_win_scale_max, - }, { .procname = "tcp_frto", .data = &sysctl_tcp_frto, @@ -1145,6 +1136,15 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_adv_win_scale", + .data = &init_net.ipv4.sysctl_tcp_adv_win_scale, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &tcp_adv_win_scale_min, + .extra2 = &tcp_adv_win_scale_max, + }, { } }; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 6af4b58ac6d5..8ee2c84b0bc6 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -79,9 +79,6 @@ #include #include -int sysctl_tcp_adv_win_scale __read_mostly = 1; -EXPORT_SYMBOL(sysctl_tcp_adv_win_scale); - /* rfc5961 challenge ack rate limiting */ int sysctl_tcp_challenge_ack_limit = 1000; @@ -363,8 +360,8 @@ static int __tcp_grow_window(const struct sock *sk, const struct sk_buff *skb) { struct tcp_sock *tp = tcp_sk(sk); /* Optimize this! */ - int truesize = tcp_win_from_space(skb->truesize) >> 1; - int window = tcp_win_from_space(sysctl_tcp_rmem[2]) >> 1; + int truesize = tcp_win_from_space(sk, skb->truesize) >> 1; + int window = tcp_win_from_space(sk, sysctl_tcp_rmem[2]) >> 1; while (tp->rcv_ssthresh <= window) { if (truesize <= skb->len) @@ -389,7 +386,7 @@ static void tcp_grow_window(struct sock *sk, const struct sk_buff *skb) /* Check #2. Increase window, if skb with such overhead * will fit to rcvbuf in future. */ - if (tcp_win_from_space(skb->truesize) <= skb->len) + if (tcp_win_from_space(sk, skb->truesize) <= skb->len) incr = 2 * tp->advmss; else incr = __tcp_grow_window(sk, skb); @@ -630,7 +627,7 @@ void tcp_rcv_space_adjust(struct sock *sk) } rcvmem = SKB_TRUESIZE(tp->advmss + MAX_TCP_HEADER); - while (tcp_win_from_space(rcvmem) < tp->advmss) + while (tcp_win_from_space(sk, rcvmem) < tp->advmss) rcvmem += 128; rcvbuf = min(rcvwin / tp->advmss * rcvmem, sysctl_tcp_rmem[2]); @@ -4809,7 +4806,7 @@ restart: * overlaps to the next one. */ if (!(TCP_SKB_CB(skb)->tcp_flags & (TCPHDR_SYN | TCPHDR_FIN)) && - (tcp_win_from_space(skb->truesize) > skb->len || + (tcp_win_from_space(sk, skb->truesize) > skb->len || before(TCP_SKB_CB(skb)->seq, start))) { end_of_skbs = false; break; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 189664ebd28e..1fe30fb99308 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2491,6 +2491,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_max_reordering = 300; net->ipv4.sysctl_tcp_dsack = 1; net->ipv4.sysctl_tcp_app_win = 31; + net->ipv4.sysctl_tcp_adv_win_scale = 1; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); -- cgit v1.2.3 From af9b69a7a6ca6b817e8d6f416e7aa5b2a5bf1d91 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 26 Oct 2017 21:55:10 -0700 Subject: tcp: Namespace-ify sysctl_tcp_frto Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_input.c | 3 +-- net/ipv4/tcp_ipv4.c | 1 + 5 files changed, 10 insertions(+), 10 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 9dbb07d4eff4..f4622e28db3a 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -141,6 +141,7 @@ struct netns_ipv4 { int sysctl_tcp_dsack; int sysctl_tcp_app_win; int sysctl_tcp_adv_win_scale; + int sysctl_tcp_frto; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 0dc27cd24899..18f047501f53 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -247,7 +247,6 @@ extern int sysctl_tcp_max_orphans; extern long sysctl_tcp_mem[3]; extern int sysctl_tcp_wmem[3]; extern int sysctl_tcp_rmem[3]; -extern int sysctl_tcp_frto; extern int sysctl_tcp_nometrics_save; extern int sysctl_tcp_moderate_rcvbuf; extern int sysctl_tcp_tso_win_divisor; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index a95123e1e7da..f1bcb9b7e082 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -437,13 +437,6 @@ static struct ctl_table ipv4_table[] = { .proc_handler = proc_dointvec_minmax, .extra1 = &one, }, - { - .procname = "tcp_frto", - .data = &sysctl_tcp_frto, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, { .procname = "tcp_min_rtt_wlen", .data = &sysctl_tcp_min_rtt_wlen, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .extra1 = &tcp_adv_win_scale_min, .extra2 = &tcp_adv_win_scale_max, }, + { + .procname = "tcp_frto", + .data = &init_net.ipv4.sysctl_tcp_frto, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 8ee2c84b0bc6..90d76f1c8f96 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -83,7 +83,6 @@ int sysctl_tcp_challenge_ack_limit = 1000; int sysctl_tcp_max_orphans __read_mostly = NR_FILE; -int sysctl_tcp_frto __read_mostly = 2; int sysctl_tcp_min_rtt_wlen __read_mostly = 300; int sysctl_tcp_moderate_rcvbuf __read_mostly = 1; int sysctl_tcp_invalid_ratelimit __read_mostly = HZ/2; @@ -2026,7 +2025,7 @@ void tcp_enter_loss(struct sock *sk) * falsely raise the receive window, which results in repeated * timeouts and stop-and-go behavior. */ - tp->frto = sysctl_tcp_frto && + tp->frto = net->ipv4.sysctl_tcp_frto && (new_recovery || icsk->icsk_retransmits) && !inet_csk(sk)->icsk_mtup.probe_size; } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 1fe30fb99308..49757c758211 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2492,6 +2492,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_dsack = 1; net->ipv4.sysctl_tcp_app_win = 31; net->ipv4.sysctl_tcp_adv_win_scale = 1; + net->ipv4.sysctl_tcp_frto = 2; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); -- cgit v1.2.3 From bff7b688d5b10a8cb8cecefdea5e255408b78f2f Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Thu, 26 Oct 2017 11:22:51 -0400 Subject: net: dsa: add dsa_is_unused_port helper As the comment above the chunk states, the b53 driver attempts to disable the unused ports. But using ds->enabled_port_mask is misleading, because this mask reports in fact the user ports. To avoid confusion and fix this, this patch introduces an explicit dsa_is_unused_port helper which ensures the corresponding bit is not masked in any of the switch port masks. Signed-off-by: Vivien Didelot Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/dsa/b53/b53_common.c | 2 +- include/net/dsa.h | 7 +++++++ 2 files changed, 8 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index b48cf0487b43..c74a50112551 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -873,7 +873,7 @@ static int b53_setup(struct dsa_switch *ds) for (port = 0; port < dev->num_ports; port++) { if (dsa_is_cpu_port(ds, port)) b53_enable_cpu_port(dev, port); - else if (!(BIT(port) & ds->enabled_port_mask)) + else if (dsa_is_unused_port(ds, port)) b53_disable_port(ds, port, NULL); } diff --git a/include/net/dsa.h b/include/net/dsa.h index 38961ef91d3d..6b1bc1c8f7e2 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -254,6 +254,13 @@ struct dsa_switch { struct dsa_port ports[]; }; +static inline bool dsa_is_unused_port(struct dsa_switch *ds, int p) +{ + u32 m = ds->enabled_port_mask | ds->dsa_port_mask | ds->cpu_port_mask; + + return !(m & BIT(p)); +} + static inline bool dsa_is_cpu_port(struct dsa_switch *ds, int p) { return !!(ds->cpu_port_mask & (1 << p)); -- cgit v1.2.3 From deb8ee0b51204273c120a3b3848efbb5695ad658 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Thu, 26 Oct 2017 11:22:53 -0400 Subject: net: dsa: fix dsa_is_normal_port helper In order to know if a port is of type user, dsa_is_normal_port checks that the given port is not of type DSA nor CPU. This is not enough because a port can be unused. Without the previous fix, this caused the unused mv88e6xxx ports to be configured in normal mode. The ds->enabled_port_mask reports the user ports, so check this instead. Signed-off-by: Vivien Didelot Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- include/net/dsa.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 6b1bc1c8f7e2..4ad432ad2d40 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -273,7 +273,7 @@ static inline bool dsa_is_dsa_port(struct dsa_switch *ds, int p) static inline bool dsa_is_normal_port(struct dsa_switch *ds, int p) { - return !dsa_is_cpu_port(ds, p) && !dsa_is_dsa_port(ds, p); + return !!(ds->enabled_port_mask & BIT(p)); } static inline const struct dsa_port *dsa_to_port(struct dsa_switch *ds, int p) -- cgit v1.2.3 From 2b3e9891cb607f7c7d5a4b11fb5a6e775e7f3ef4 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Thu, 26 Oct 2017 11:22:54 -0400 Subject: net: dsa: rename dsa_is_normal_port helper This patch renames dsa_is_normal_port to dsa_is_user_port because "user" is the correct term in the DSA terminology, not "normal". Signed-off-by: Vivien Didelot Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/dsa/mv88e6xxx/chip.c | 2 +- include/net/dsa.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index 2d8cf66e8f74..09a66d4d9492 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -1676,7 +1676,7 @@ static int mv88e6xxx_setup_port_mode(struct mv88e6xxx_chip *chip, int port) if (dsa_is_dsa_port(chip->ds, port)) return mv88e6xxx_set_port_mode_dsa(chip, port); - if (dsa_is_normal_port(chip->ds, port)) + if (dsa_is_user_port(chip->ds, port)) return mv88e6xxx_set_port_mode_normal(chip, port); /* Setup CPU port mode depending on its supported tag format */ diff --git a/include/net/dsa.h b/include/net/dsa.h index 4ad432ad2d40..49701d958663 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -271,7 +271,7 @@ static inline bool dsa_is_dsa_port(struct dsa_switch *ds, int p) return !!((ds->dsa_port_mask) & (1 << p)); } -static inline bool dsa_is_normal_port(struct dsa_switch *ds, int p) +static inline bool dsa_is_user_port(struct dsa_switch *ds, int p) { return !!(ds->enabled_port_mask & BIT(p)); } -- cgit v1.2.3 From 02bc6e546e858b209c3ebe380a13a73b333b1b3f Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Thu, 26 Oct 2017 11:22:56 -0400 Subject: net: dsa: introduce dsa_user_ports helper Introduce a dsa_user_ports() helper to return the ds->enabled_port_mask mask which is more explicit. This will also minimize diffs when touching this internal mask. Signed-off-by: Vivien Didelot Signed-off-by: David S. Miller --- drivers/net/dsa/mt7530.c | 2 +- drivers/net/dsa/mv88e6060.c | 5 ++--- drivers/net/dsa/qca8k.c | 3 +-- include/net/dsa.h | 5 +++++ net/dsa/dsa2.c | 2 +- net/dsa/legacy.c | 2 +- 6 files changed, 11 insertions(+), 8 deletions(-) (limited to 'include/net') diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c index d1d4eea6a875..627c039f12ca 100644 --- a/drivers/net/dsa/mt7530.c +++ b/drivers/net/dsa/mt7530.c @@ -688,7 +688,7 @@ mt7530_cpu_port_enable(struct mt7530_priv *priv, * the switch */ mt7530_write(priv, MT7530_PCR_P(port), - PCR_MATRIX(priv->ds->enabled_port_mask)); + PCR_MATRIX(dsa_user_ports(priv->ds))); return 0; } diff --git a/drivers/net/dsa/mv88e6060.c b/drivers/net/dsa/mv88e6060.c index f78b9e13be1c..45768e3c5bc5 100644 --- a/drivers/net/dsa/mv88e6060.c +++ b/drivers/net/dsa/mv88e6060.c @@ -175,9 +175,8 @@ static int mv88e6060_setup_port(struct dsa_switch *ds, int p) */ REG_WRITE(addr, PORT_VLAN_MAP, ((p & 0xf) << PORT_VLAN_MAP_DBNUM_SHIFT) | - (dsa_is_cpu_port(ds, p) ? - ds->enabled_port_mask : - BIT(dsa_to_port(ds, p)->cpu_dp->index))); + (dsa_is_cpu_port(ds, p) ? dsa_user_ports(ds) : + BIT(dsa_to_port(ds, p)->cpu_dp->index))); /* Port Association Vector: when learning source addresses * of packets, add the address to the address database using diff --git a/drivers/net/dsa/qca8k.c b/drivers/net/dsa/qca8k.c index 37125858fe80..cf72e274275f 100644 --- a/drivers/net/dsa/qca8k.c +++ b/drivers/net/dsa/qca8k.c @@ -551,8 +551,7 @@ qca8k_setup(struct dsa_switch *ds) /* CPU port gets connected to all user ports of the switch */ if (dsa_is_cpu_port(ds, i)) { qca8k_rmw(priv, QCA8K_PORT_LOOKUP_CTRL(QCA8K_CPU_PORT), - QCA8K_PORT_LOOKUP_MEMBER, - ds->enabled_port_mask); + QCA8K_PORT_LOOKUP_MEMBER, dsa_user_ports(ds)); } /* Invividual user ports get connected to CPU port only */ diff --git a/include/net/dsa.h b/include/net/dsa.h index 49701d958663..dc7728062396 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -276,6 +276,11 @@ static inline bool dsa_is_user_port(struct dsa_switch *ds, int p) return !!(ds->enabled_port_mask & BIT(p)); } +static inline u32 dsa_user_ports(struct dsa_switch *ds) +{ + return ds->enabled_port_mask; +} + static inline const struct dsa_port *dsa_to_port(struct dsa_switch *ds, int p) { return &ds->ports[p]; diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index 62485a57dbfc..d43c59c91058 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -312,7 +312,7 @@ static int dsa_ds_apply(struct dsa_switch_tree *dst, struct dsa_switch *ds) * the slave MDIO bus driver rely on these values for probing PHY * devices or not */ - ds->phys_mii_mask = ds->enabled_port_mask; + ds->phys_mii_mask |= dsa_user_ports(ds); /* Add the switch to devlink before calling setup, so that setup can * add dpipe tables diff --git a/net/dsa/legacy.c b/net/dsa/legacy.c index 0b79c6171d0d..fa543c4a6061 100644 --- a/net/dsa/legacy.c +++ b/net/dsa/legacy.c @@ -136,7 +136,7 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, /* Make the built-in MII bus mask match the number of ports, * switch drivers can override this later */ - ds->phys_mii_mask = ds->enabled_port_mask; + ds->phys_mii_mask |= dsa_user_ports(ds); /* * If the CPU connects to this switch, set the switch tree -- cgit v1.2.3 From 057cad2c59d73b0c4a6638546f3099d6fb444094 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Thu, 26 Oct 2017 11:22:57 -0400 Subject: net: dsa: define port types Introduce an enumerated type for ports, which will be way more explicit to identify a port type instead of digging into switch port masks. A port can be of type CPU, DSA, user, or unused by default. This is a static parsed information that cannot be changed at runtime. Signed-off-by: Vivien Didelot Signed-off-by: David S. Miller --- include/net/dsa.h | 7 +++++++ net/dsa/dsa2.c | 3 +++ net/dsa/legacy.c | 6 ++++++ 3 files changed, 16 insertions(+) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index dc7728062396..8da20c4a6552 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -180,6 +180,13 @@ struct dsa_port { struct sk_buff *(*rcv)(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt); + enum { + DSA_PORT_TYPE_UNUSED = 0, + DSA_PORT_TYPE_CPU, + DSA_PORT_TYPE_DSA, + DSA_PORT_TYPE_USER, + } type; + struct dsa_switch *ds; unsigned int index; const char *name; diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index d43c59c91058..dd6f35b92937 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -185,6 +185,7 @@ static int dsa_ds_complete(struct dsa_switch_tree *dst, struct dsa_switch *ds) return err; ds->dsa_port_mask |= BIT(index); + port->type = DSA_PORT_TYPE_DSA; } return 0; @@ -504,6 +505,7 @@ static int dsa_cpu_parse(struct dsa_port *port, u32 index, * net/dsa/dsa.c::dsa_switch_setup_one does. */ ds->cpu_port_mask |= BIT(index); + port->type = DSA_PORT_TYPE_CPU; tag_protocol = ds->ops->get_tag_protocol(ds); tag_ops = dsa_resolve_tag_protocol(tag_protocol); @@ -543,6 +545,7 @@ static int dsa_ds_parse(struct dsa_switch_tree *dst, struct dsa_switch *ds) * net/dsa/dsa.c::dsa_switch_setup_one does. */ ds->enabled_port_mask |= BIT(index); + port->type = DSA_PORT_TYPE_USER; } } diff --git a/net/dsa/legacy.c b/net/dsa/legacy.c index fa543c4a6061..9fd5b3adab1e 100644 --- a/net/dsa/legacy.c +++ b/net/dsa/legacy.c @@ -101,6 +101,7 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, struct dsa_chip_data *cd = ds->cd; bool valid_name_found = false; int index = ds->index; + struct dsa_port *dp; int i, ret; /* @@ -109,6 +110,8 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, for (i = 0; i < ds->num_ports; i++) { char *name; + dp = &ds->ports[i]; + name = cd->port_names[i]; if (name == NULL) continue; @@ -122,10 +125,13 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, dst->cpu_dp = &ds->ports[i]; dst->cpu_dp->master = master; ds->cpu_port_mask |= 1 << i; + dp->type = DSA_PORT_TYPE_CPU; } else if (!strcmp(name, "dsa")) { ds->dsa_port_mask |= 1 << i; + dp->type = DSA_PORT_TYPE_DSA; } else { ds->enabled_port_mask |= 1 << i; + dp->type = DSA_PORT_TYPE_USER; } valid_name_found = true; } -- cgit v1.2.3 From c38c5a66506e4e8223fd03e950b1bde99190701e Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Thu, 26 Oct 2017 11:22:58 -0400 Subject: net: dsa: use new port type in helpers Now that DSA exposes an enumerated type for the ports, we can use them directly instead of checking bitmaps, which is more consistent. Signed-off-by: Vivien Didelot Signed-off-by: David S. Miller --- include/net/dsa.h | 27 ++++++++++++++++----------- 1 file changed, 16 insertions(+), 11 deletions(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 8da20c4a6552..07dfbd7f4fd5 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -261,36 +261,41 @@ struct dsa_switch { struct dsa_port ports[]; }; -static inline bool dsa_is_unused_port(struct dsa_switch *ds, int p) +static inline const struct dsa_port *dsa_to_port(struct dsa_switch *ds, int p) { - u32 m = ds->enabled_port_mask | ds->dsa_port_mask | ds->cpu_port_mask; + return &ds->ports[p]; +} - return !(m & BIT(p)); +static inline bool dsa_is_unused_port(struct dsa_switch *ds, int p) +{ + return dsa_to_port(ds, p)->type == DSA_PORT_TYPE_UNUSED; } static inline bool dsa_is_cpu_port(struct dsa_switch *ds, int p) { - return !!(ds->cpu_port_mask & (1 << p)); + return dsa_to_port(ds, p)->type == DSA_PORT_TYPE_CPU; } static inline bool dsa_is_dsa_port(struct dsa_switch *ds, int p) { - return !!((ds->dsa_port_mask) & (1 << p)); + return dsa_to_port(ds, p)->type == DSA_PORT_TYPE_DSA; } static inline bool dsa_is_user_port(struct dsa_switch *ds, int p) { - return !!(ds->enabled_port_mask & BIT(p)); + return dsa_to_port(ds, p)->type == DSA_PORT_TYPE_USER; } static inline u32 dsa_user_ports(struct dsa_switch *ds) { - return ds->enabled_port_mask; -} + u32 mask = 0; + int p; -static inline const struct dsa_port *dsa_to_port(struct dsa_switch *ds, int p) -{ - return &ds->ports[p]; + for (p = 0; p < ds->num_ports; p++) + if (dsa_is_user_port(ds, p)) + mask |= BIT(p); + + return mask; } static inline u8 dsa_upstream_port(struct dsa_switch *ds) -- cgit v1.2.3 From 5749f0f3772b9d98f37e3a92539f49fafaa64eca Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Thu, 26 Oct 2017 11:22:59 -0400 Subject: net: dsa: remove port masks Now that DSA core provides port types, there is no need to keep this information at the switch level. This is a static information that is part of a DSA core dsa_port structure. Remove them. Signed-off-by: Vivien Didelot Signed-off-by: David S. Miller --- include/net/dsa.h | 3 --- net/dsa/dsa2.c | 11 ----------- net/dsa/legacy.c | 3 --- 3 files changed, 17 deletions(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 07dfbd7f4fd5..50e276dc4c01 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -240,9 +240,6 @@ struct dsa_switch { /* * Slave mii_bus and devices for the individual ports. */ - u32 dsa_port_mask; - u32 cpu_port_mask; - u32 enabled_port_mask; u32 phys_mii_mask; struct mii_bus *slave_mii_bus; diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index dd6f35b92937..ec58654a71cd 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -184,7 +184,6 @@ static int dsa_ds_complete(struct dsa_switch_tree *dst, struct dsa_switch *ds) if (err != 0) return err; - ds->dsa_port_mask |= BIT(index); port->type = DSA_PORT_TYPE_DSA; } @@ -500,11 +499,6 @@ static int dsa_cpu_parse(struct dsa_port *port, u32 index, dst->cpu_dp->master = ethernet_dev; } - /* Initialize cpu_port_mask now for drv->setup() - * to have access to a correct value, just like what - * net/dsa/dsa.c::dsa_switch_setup_one does. - */ - ds->cpu_port_mask |= BIT(index); port->type = DSA_PORT_TYPE_CPU; tag_protocol = ds->ops->get_tag_protocol(ds); @@ -540,11 +534,6 @@ static int dsa_ds_parse(struct dsa_switch_tree *dst, struct dsa_switch *ds) if (err) return err; } else { - /* Initialize enabled_port_mask now for drv->setup() - * to have access to a correct value, just like what - * net/dsa/dsa.c::dsa_switch_setup_one does. - */ - ds->enabled_port_mask |= BIT(index); port->type = DSA_PORT_TYPE_USER; } diff --git a/net/dsa/legacy.c b/net/dsa/legacy.c index 9fd5b3adab1e..93e1b116ef83 100644 --- a/net/dsa/legacy.c +++ b/net/dsa/legacy.c @@ -124,13 +124,10 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, } dst->cpu_dp = &ds->ports[i]; dst->cpu_dp->master = master; - ds->cpu_port_mask |= 1 << i; dp->type = DSA_PORT_TYPE_CPU; } else if (!strcmp(name, "dsa")) { - ds->dsa_port_mask |= 1 << i; dp->type = DSA_PORT_TYPE_DSA; } else { - ds->enabled_port_mask |= 1 << i; dp->type = DSA_PORT_TYPE_USER; } valid_name_found = true; -- cgit v1.2.3 From 3d0bd028ffb4a4915cb64cfa0d2cee1578cc0321 Mon Sep 17 00:00:00 2001 From: Vinicius Costa Gomes Date: Mon, 16 Oct 2017 18:01:27 -0700 Subject: net/sched: Add support for HW offloading for CBS This adds support for offloading the CBS algorithm to the controller, if supported. Drivers wanting to support CBS offload must implement the .ndo_setup_tc callback and handle the TC_SETUP_CBS (introduced here) type. Signed-off-by: Vinicius Costa Gomes Tested-by: Henrik Austad Signed-off-by: Jeff Kirsher --- include/linux/netdevice.h | 1 + include/net/pkt_sched.h | 9 ++++ net/sched/sch_cbs.c | 104 ++++++++++++++++++++++++++++++++++++++++------ 3 files changed, 102 insertions(+), 12 deletions(-) (limited to 'include/net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 6c7960c8338a..5e02f79b2110 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -776,6 +776,7 @@ enum tc_setup_type { TC_SETUP_CLSMATCHALL, TC_SETUP_CLSBPF, TC_SETUP_BLOCK, + TC_SETUP_CBS, }; /* These structures hold the attributes of xdp state that are being passed diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index b8ecafce4ba1..02f2db26e30c 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -140,4 +140,13 @@ static inline struct net *qdisc_net(struct Qdisc *q) return dev_net(q->dev_queue->dev); } +struct tc_cbs_qopt_offload { + u8 enable; + s32 queue; + s32 hicredit; + s32 locredit; + s32 idleslope; + s32 sendslope; +}; + #endif diff --git a/net/sched/sch_cbs.c b/net/sched/sch_cbs.c index 0e85133c5653..bdb533b7fb8c 100644 --- a/net/sched/sch_cbs.c +++ b/net/sched/sch_cbs.c @@ -68,6 +68,8 @@ #define BYTES_PER_KBIT (1000LL / 8) struct cbs_sched_data { + bool offload; + int queue; s64 port_rate; /* in bytes/s */ s64 last; /* timestamp in ns */ s64 credits; /* in bytes */ @@ -80,6 +82,11 @@ struct cbs_sched_data { struct sk_buff *(*dequeue)(struct Qdisc *sch); }; +static int cbs_enqueue_offload(struct sk_buff *skb, struct Qdisc *sch) +{ + return qdisc_enqueue_tail(skb, sch); +} + static int cbs_enqueue_soft(struct sk_buff *skb, struct Qdisc *sch) { struct cbs_sched_data *q = qdisc_priv(sch); @@ -169,6 +176,11 @@ static struct sk_buff *cbs_dequeue_soft(struct Qdisc *sch) return skb; } +static struct sk_buff *cbs_dequeue_offload(struct Qdisc *sch) +{ + return qdisc_dequeue_head(sch); +} + static struct sk_buff *cbs_dequeue(struct Qdisc *sch) { struct cbs_sched_data *q = qdisc_priv(sch); @@ -180,14 +192,66 @@ static const struct nla_policy cbs_policy[TCA_CBS_MAX + 1] = { [TCA_CBS_PARMS] = { .len = sizeof(struct tc_cbs_qopt) }, }; +static void cbs_disable_offload(struct net_device *dev, + struct cbs_sched_data *q) +{ + struct tc_cbs_qopt_offload cbs = { }; + const struct net_device_ops *ops; + int err; + + if (!q->offload) + return; + + q->enqueue = cbs_enqueue_soft; + q->dequeue = cbs_dequeue_soft; + + ops = dev->netdev_ops; + if (!ops->ndo_setup_tc) + return; + + cbs.queue = q->queue; + cbs.enable = 0; + + err = ops->ndo_setup_tc(dev, TC_SETUP_CBS, &cbs); + if (err < 0) + pr_warn("Couldn't disable CBS offload for queue %d\n", + cbs.queue); +} + +static int cbs_enable_offload(struct net_device *dev, struct cbs_sched_data *q, + const struct tc_cbs_qopt *opt) +{ + const struct net_device_ops *ops = dev->netdev_ops; + struct tc_cbs_qopt_offload cbs = { }; + int err; + + if (!ops->ndo_setup_tc) + return -EOPNOTSUPP; + + cbs.queue = q->queue; + + cbs.enable = 1; + cbs.hicredit = opt->hicredit; + cbs.locredit = opt->locredit; + cbs.idleslope = opt->idleslope; + cbs.sendslope = opt->sendslope; + + err = ops->ndo_setup_tc(dev, TC_SETUP_CBS, &cbs); + if (err < 0) + return err; + + q->enqueue = cbs_enqueue_offload; + q->dequeue = cbs_dequeue_offload; + + return 0; +} + static int cbs_change(struct Qdisc *sch, struct nlattr *opt) { struct cbs_sched_data *q = qdisc_priv(sch); struct net_device *dev = qdisc_dev(sch); struct nlattr *tb[TCA_CBS_MAX + 1]; - struct ethtool_link_ksettings ecmd; struct tc_cbs_qopt *qopt; - s64 link_speed; int err; err = nla_parse_nested(tb, TCA_CBS_MAX, opt, cbs_policy, NULL); @@ -199,23 +263,30 @@ static int cbs_change(struct Qdisc *sch, struct nlattr *opt) qopt = nla_data(tb[TCA_CBS_PARMS]); - if (qopt->offload) - return -EOPNOTSUPP; + if (!qopt->offload) { + struct ethtool_link_ksettings ecmd; + s64 link_speed; - if (!__ethtool_get_link_ksettings(dev, &ecmd)) - link_speed = ecmd.base.speed; - else - link_speed = SPEED_1000; + if (!__ethtool_get_link_ksettings(dev, &ecmd)) + link_speed = ecmd.base.speed; + else + link_speed = SPEED_1000; - q->port_rate = link_speed * 1000 * BYTES_PER_KBIT; + q->port_rate = link_speed * 1000 * BYTES_PER_KBIT; - q->enqueue = cbs_enqueue_soft; - q->dequeue = cbs_dequeue_soft; + cbs_disable_offload(dev, q); + } else { + err = cbs_enable_offload(dev, q, qopt); + if (err < 0) + return err; + } + /* Everything went OK, save the parameters used. */ q->hicredit = qopt->hicredit; q->locredit = qopt->locredit; q->idleslope = qopt->idleslope * BYTES_PER_KBIT; q->sendslope = qopt->sendslope * BYTES_PER_KBIT; + q->offload = qopt->offload; return 0; } @@ -223,10 +294,16 @@ static int cbs_change(struct Qdisc *sch, struct nlattr *opt) static int cbs_init(struct Qdisc *sch, struct nlattr *opt) { struct cbs_sched_data *q = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); if (!opt) return -EINVAL; + q->queue = sch->dev_queue - netdev_get_tx_queue(dev, 0); + + q->enqueue = cbs_enqueue_soft; + q->dequeue = cbs_dequeue_soft; + qdisc_watchdog_init(&q->watchdog, sch); return cbs_change(sch, opt); @@ -235,8 +312,11 @@ static int cbs_init(struct Qdisc *sch, struct nlattr *opt) static void cbs_destroy(struct Qdisc *sch) { struct cbs_sched_data *q = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); qdisc_watchdog_cancel(&q->watchdog); + + cbs_disable_offload(dev, q); } static int cbs_dump(struct Qdisc *sch, struct sk_buff *skb) @@ -253,7 +333,7 @@ static int cbs_dump(struct Qdisc *sch, struct sk_buff *skb) opt.locredit = q->locredit; opt.sendslope = div64_s64(q->sendslope, BYTES_PER_KBIT); opt.idleslope = div64_s64(q->idleslope, BYTES_PER_KBIT); - opt.offload = 0; + opt.offload = q->offload; if (nla_put(skb, TCA_CBS_PARMS, sizeof(opt), &opt)) goto nla_put_failure; -- cgit v1.2.3 From ec36e416f06f6a8659281053fdc46ce484ad2211 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 27 Oct 2017 07:47:21 -0700 Subject: tcp: Namespace-ify sysctl_tcp_nometrics_save Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_metrics.c | 4 +--- 4 files changed, 9 insertions(+), 11 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index f4622e28db3a..9606e2ea1f14 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -142,6 +142,7 @@ struct netns_ipv4 { int sysctl_tcp_app_win; int sysctl_tcp_adv_win_scale; int sysctl_tcp_frto; + int sysctl_tcp_nometrics_save; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 18f047501f53..6ab7fa4154b2 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -247,7 +247,6 @@ extern int sysctl_tcp_max_orphans; extern long sysctl_tcp_mem[3]; extern int sysctl_tcp_wmem[3]; extern int sysctl_tcp_rmem[3]; -extern int sysctl_tcp_nometrics_save; extern int sysctl_tcp_moderate_rcvbuf; extern int sysctl_tcp_tso_win_divisor; extern int sysctl_tcp_workaround_signed_windows; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index f1bcb9b7e082..b742a5e26a9d 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -451,13 +451,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "tcp_no_metrics_save", - .data = &sysctl_tcp_nometrics_save, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec, - }, { .procname = "tcp_moderate_rcvbuf", .data = &sysctl_tcp_moderate_rcvbuf, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_no_metrics_save", + .data = &init_net.ipv4.sysctl_tcp_nometrics_save, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, { } }; diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp_metrics.c index 0ab78abc811b..0507b56b6d4b 100644 --- a/net/ipv4/tcp_metrics.c +++ b/net/ipv4/tcp_metrics.c @@ -20,8 +20,6 @@ #include #include -int sysctl_tcp_nometrics_save __read_mostly; - static struct tcp_metrics_block *__tcp_get_metrics(const struct inetpeer_addr *saddr, const struct inetpeer_addr *daddr, struct net *net, unsigned int hash); @@ -330,7 +328,7 @@ void tcp_update_metrics(struct sock *sk) int m; sk_dst_confirm(sk); - if (sysctl_tcp_nometrics_save || !dst) + if (net->ipv4.sysctl_tcp_nometrics_save || !dst) return; rcu_read_lock(); -- cgit v1.2.3 From 4540c0cf98b8892a642d2453eec20ae3eb5696fb Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 27 Oct 2017 07:47:22 -0700 Subject: tcp: Namespace-ify sysctl_tcp_moderate_rcvbuf Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_input.c | 5 ++--- net/ipv4/tcp_ipv4.c | 1 + 5 files changed, 11 insertions(+), 11 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 9606e2ea1f14..4458a54fe3f4 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -143,6 +143,7 @@ struct netns_ipv4 { int sysctl_tcp_adv_win_scale; int sysctl_tcp_frto; int sysctl_tcp_nometrics_save; + int sysctl_tcp_moderate_rcvbuf; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 6ab7fa4154b2..f954e74578ff 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -247,7 +247,6 @@ extern int sysctl_tcp_max_orphans; extern long sysctl_tcp_mem[3]; extern int sysctl_tcp_wmem[3]; extern int sysctl_tcp_rmem[3]; -extern int sysctl_tcp_moderate_rcvbuf; extern int sysctl_tcp_tso_win_divisor; extern int sysctl_tcp_workaround_signed_windows; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index b742a5e26a9d..2ebe87fd1169 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -451,13 +451,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "tcp_moderate_rcvbuf", - .data = &sysctl_tcp_moderate_rcvbuf, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec, - }, { .procname = "tcp_tso_win_divisor", .data = &sysctl_tcp_tso_win_divisor, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, + { + .procname = "tcp_moderate_rcvbuf", + .data = &init_net.ipv4.sysctl_tcp_moderate_rcvbuf, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, { } }; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 90d76f1c8f96..ce481325115f 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -84,7 +84,6 @@ int sysctl_tcp_challenge_ack_limit = 1000; int sysctl_tcp_max_orphans __read_mostly = NR_FILE; int sysctl_tcp_min_rtt_wlen __read_mostly = 300; -int sysctl_tcp_moderate_rcvbuf __read_mostly = 1; int sysctl_tcp_invalid_ratelimit __read_mostly = HZ/2; #define FLAG_DATA 0x01 /* Incoming frame contained data. */ @@ -411,7 +410,7 @@ static void tcp_fixup_rcvbuf(struct sock *sk) /* Dynamic Right Sizing (DRS) has 2 to 3 RTT latency * Allow enough cushion so that sender is not limited by our window */ - if (sysctl_tcp_moderate_rcvbuf) + if (sock_net(sk)->ipv4.sysctl_tcp_moderate_rcvbuf) rcvmem <<= 2; if (sk->sk_rcvbuf < rcvmem) @@ -602,7 +601,7 @@ void tcp_rcv_space_adjust(struct sock *sk) * */ - if (sysctl_tcp_moderate_rcvbuf && + if (sock_net(sk)->ipv4.sysctl_tcp_moderate_rcvbuf && !(sk->sk_userlocks & SOCK_RCVBUF_LOCK)) { int rcvwin, rcvmem, rcvbuf; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 49757c758211..27f376b90913 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2493,6 +2493,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_app_win = 31; net->ipv4.sysctl_tcp_adv_win_scale = 1; net->ipv4.sysctl_tcp_frto = 2; + net->ipv4.sysctl_tcp_moderate_rcvbuf = 1; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); -- cgit v1.2.3 From d06a99045837d3f4d5431793c4c390b0daf2a08d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 27 Oct 2017 07:47:23 -0700 Subject: tcp: Namespace-ify sysctl_tcp_tso_win_divisor Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_ipv4.c | 5 +++++ net/ipv4/tcp_output.c | 8 +------- 5 files changed, 14 insertions(+), 15 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 4458a54fe3f4..60bccda046db 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -144,6 +144,7 @@ struct netns_ipv4 { int sysctl_tcp_frto; int sysctl_tcp_nometrics_save; int sysctl_tcp_moderate_rcvbuf; + int sysctl_tcp_tso_win_divisor; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index f954e74578ff..ed0828dc82f1 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -247,7 +247,6 @@ extern int sysctl_tcp_max_orphans; extern long sysctl_tcp_mem[3]; extern int sysctl_tcp_wmem[3]; extern int sysctl_tcp_rmem[3]; -extern int sysctl_tcp_tso_win_divisor; extern int sysctl_tcp_workaround_signed_windows; #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 2ebe87fd1169..a053cacb8290 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -451,13 +451,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "tcp_tso_win_divisor", - .data = &sysctl_tcp_tso_win_divisor, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec, - }, { .procname = "tcp_congestion_control", .mode = 0644, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, + { + .procname = "tcp_tso_win_divisor", + .data = &init_net.ipv4.sysctl_tcp_tso_win_divisor, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, { } }; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 27f376b90913..284ff16148df 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2494,6 +2494,11 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_adv_win_scale = 1; net->ipv4.sysctl_tcp_frto = 2; net->ipv4.sysctl_tcp_moderate_rcvbuf = 1; + /* This limits the percentage of the congestion window which we + * will allow a single TSO frame to consume. Building TSO frames + * which are too large can cause TCP streams to be bursty. + */ + net->ipv4.sysctl_tcp_tso_win_divisor = 3; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 55a0aa4b96df..60df3ab52166 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -53,12 +53,6 @@ int sysctl_tcp_workaround_signed_windows __read_mostly = 0; /* Default TSQ limit of four TSO segments */ int sysctl_tcp_limit_output_bytes __read_mostly = 262144; -/* This limits the percentage of the congestion window which we - * will allow a single TSO frame to consume. Building TSO frames - * which are too large can cause TCP streams to be bursty. - */ -int sysctl_tcp_tso_win_divisor __read_mostly = 3; - static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, int push_one, gfp_t gfp); @@ -1988,7 +1982,7 @@ static bool tcp_tso_should_defer(struct sock *sk, struct sk_buff *skb, if ((skb != tcp_write_queue_tail(sk)) && (limit >= skb->len)) goto send_now; - win_divisor = ACCESS_ONCE(sysctl_tcp_tso_win_divisor); + win_divisor = ACCESS_ONCE(sock_net(sk)->ipv4.sysctl_tcp_tso_win_divisor); if (win_divisor) { u32 chunk = min(tp->snd_wnd, tp->snd_cwnd * tp->mss_cache); -- cgit v1.2.3 From ceef9ab6be7234f9e49f79769e0da88d1dccfcc7 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 27 Oct 2017 07:47:24 -0700 Subject: tcp: Namespace-ify sysctl_tcp_workaround_signed_windows Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 4 ++-- net/ipv4/syncookies.c | 2 +- net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_minisocks.c | 2 +- net/ipv4/tcp_output.c | 14 +++++--------- net/ipv6/syncookies.c | 2 +- 7 files changed, 18 insertions(+), 21 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 60bccda046db..e74c7c1b0d18 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -145,6 +145,7 @@ struct netns_ipv4 { int sysctl_tcp_nometrics_save; int sysctl_tcp_moderate_rcvbuf; int sysctl_tcp_tso_win_divisor; + int sysctl_tcp_workaround_signed_windows; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index ed0828dc82f1..e338e16178dd 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -247,7 +247,6 @@ extern int sysctl_tcp_max_orphans; extern long sysctl_tcp_mem[3]; extern int sysctl_tcp_wmem[3]; extern int sysctl_tcp_rmem[3]; -extern int sysctl_tcp_workaround_signed_windows; #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ @@ -1302,7 +1301,8 @@ static inline void tcp_slow_start_after_idle_check(struct sock *sk) } /* Determine a window scaling and initial window to offer. */ -void tcp_select_initial_window(int __space, __u32 mss, __u32 *rcv_wnd, +void tcp_select_initial_window(const struct sock *sk, int __space, + __u32 mss, __u32 *rcv_wnd, __u32 *window_clamp, int wscale_ok, __u8 *rcv_wscale, __u32 init_rcv_wnd); diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index 77cf32a80952..fda37f2862c9 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -385,7 +385,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb) /* Try to redo what tcp_v4_send_synack did. */ req->rsk_window_clamp = tp->window_clamp ? :dst_metric(&rt->dst, RTAX_WINDOW); - tcp_select_initial_window(tcp_full_space(sk), req->mss, + tcp_select_initial_window(sk, tcp_full_space(sk), req->mss, &req->rsk_rcv_wnd, &req->rsk_window_clamp, ireq->wscale_ok, &rcv_wscale, dst_metric(&rt->dst, RTAX_INITRWND)); diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index a053cacb8290..3ae9012a4979 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -457,13 +457,6 @@ static struct ctl_table ipv4_table[] = { .maxlen = TCP_CA_NAME_MAX, .proc_handler = proc_tcp_congestion_control, }, - { - .procname = "tcp_workaround_signed_windows", - .data = &sysctl_tcp_workaround_signed_windows, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, { .procname = "tcp_limit_output_bytes", .data = &sysctl_tcp_limit_output_bytes, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec, }, + { + .procname = "tcp_workaround_signed_windows", + .data = &init_net.ipv4.sysctl_tcp_workaround_signed_windows, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 3270ab8416ce..3c65c1a3f944 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -369,7 +369,7 @@ void tcp_openreq_init_rwin(struct request_sock *req, full_space = rcv_wnd * mss; /* tcp_full_space because it is guaranteed to be the first packet */ - tcp_select_initial_window(full_space, + tcp_select_initial_window(sk_listener, full_space, mss - (ireq->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0), &req->rsk_rcv_wnd, &req->rsk_window_clamp, diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 60df3ab52166..5bbed67c27e9 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -45,11 +45,6 @@ #include -/* People can turn this on to work with those rare, broken TCPs that - * interpret the window field as a signed quantity. - */ -int sysctl_tcp_workaround_signed_windows __read_mostly = 0; - /* Default TSQ limit of four TSO segments */ int sysctl_tcp_limit_output_bytes __read_mostly = 262144; @@ -196,7 +191,7 @@ u32 tcp_default_init_rwnd(u32 mss) * be a multiple of mss if possible. We assume here that mss >= 1. * This MUST be enforced by all callers. */ -void tcp_select_initial_window(int __space, __u32 mss, +void tcp_select_initial_window(const struct sock *sk, int __space, __u32 mss, __u32 *rcv_wnd, __u32 *window_clamp, int wscale_ok, __u8 *rcv_wscale, __u32 init_rcv_wnd) @@ -220,7 +215,7 @@ void tcp_select_initial_window(int __space, __u32 mss, * which we interpret as a sign the remote TCP is not * misinterpreting the window field as a signed quantity. */ - if (sysctl_tcp_workaround_signed_windows) + if (sock_net(sk)->ipv4.sysctl_tcp_workaround_signed_windows) (*rcv_wnd) = min(space, MAX_TCP_WINDOW); else (*rcv_wnd) = space; @@ -280,7 +275,8 @@ static u16 tcp_select_window(struct sock *sk) /* Make sure we do not exceed the maximum possible * scaled window. */ - if (!tp->rx_opt.rcv_wscale && sysctl_tcp_workaround_signed_windows) + if (!tp->rx_opt.rcv_wscale && + sock_net(sk)->ipv4.sysctl_tcp_workaround_signed_windows) new_win = min(new_win, MAX_TCP_WINDOW); else new_win = min(new_win, (65535U << tp->rx_opt.rcv_wscale)); @@ -3349,7 +3345,7 @@ static void tcp_connect_init(struct sock *sk) if (rcv_wnd == 0) rcv_wnd = dst_metric(dst, RTAX_INITRWND); - tcp_select_initial_window(tcp_full_space(sk), + tcp_select_initial_window(sk, tcp_full_space(sk), tp->advmss - (tp->rx_opt.ts_recent_stamp ? tp->tcp_header_len - sizeof(struct tcphdr) : 0), &tp->rcv_wnd, &tp->window_clamp, diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c index 4e7817abc0b9..e7a3a6b6cf56 100644 --- a/net/ipv6/syncookies.c +++ b/net/ipv6/syncookies.c @@ -244,7 +244,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb) } req->rsk_window_clamp = tp->window_clamp ? :dst_metric(dst, RTAX_WINDOW); - tcp_select_initial_window(tcp_full_space(sk), req->mss, + tcp_select_initial_window(sk, tcp_full_space(sk), req->mss, &req->rsk_rcv_wnd, &req->rsk_window_clamp, ireq->wscale_ok, &rcv_wscale, dst_metric(dst, RTAX_INITRWND)); -- cgit v1.2.3 From 9184d8bb448a3d2c2d9f90f1e2f5de625292e769 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 27 Oct 2017 07:47:25 -0700 Subject: tcp: Namespace-ify sysctl_tcp_limit_output_bytes Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_ipv4.c | 2 ++ net/ipv4/tcp_output.c | 6 ++---- 5 files changed, 12 insertions(+), 12 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index e74c7c1b0d18..e98f473bab13 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -146,6 +146,7 @@ struct netns_ipv4 { int sysctl_tcp_moderate_rcvbuf; int sysctl_tcp_tso_win_divisor; int sysctl_tcp_workaround_signed_windows; + int sysctl_tcp_limit_output_bytes; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index e338e16178dd..33f9d30a6905 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -250,7 +250,6 @@ extern int sysctl_tcp_rmem[3]; #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ -extern int sysctl_tcp_limit_output_bytes; extern int sysctl_tcp_challenge_ack_limit; extern int sysctl_tcp_min_tso_segs; extern int sysctl_tcp_min_rtt_wlen; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 3ae9012a4979..6caf5c40730f 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -457,13 +457,6 @@ static struct ctl_table ipv4_table[] = { .maxlen = TCP_CA_NAME_MAX, .proc_handler = proc_tcp_congestion_control, }, - { - .procname = "tcp_limit_output_bytes", - .data = &sysctl_tcp_limit_output_bytes, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, { .procname = "tcp_challenge_ack_limit", .data = &sysctl_tcp_challenge_ack_limit, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_limit_output_bytes", + .data = &init_net.ipv4.sysctl_tcp_limit_output_bytes, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 284ff16148df..713b80261e4f 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2499,6 +2499,8 @@ static int __net_init tcp_sk_init(struct net *net) * which are too large can cause TCP streams to be bursty. */ net->ipv4.sysctl_tcp_tso_win_divisor = 3; + /* Default TSQ limit of four TSO segments */ + net->ipv4.sysctl_tcp_limit_output_bytes = 262144; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 5bbed67c27e9..f018892c6a98 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -45,9 +45,6 @@ #include -/* Default TSQ limit of four TSO segments */ -int sysctl_tcp_limit_output_bytes __read_mostly = 262144; - static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, int push_one, gfp_t gfp); @@ -2215,7 +2212,8 @@ static bool tcp_small_queue_check(struct sock *sk, const struct sk_buff *skb, unsigned int limit; limit = max(2 * skb->truesize, sk->sk_pacing_rate >> 10); - limit = min_t(u32, limit, sysctl_tcp_limit_output_bytes); + limit = min_t(u32, limit, + sock_net(sk)->ipv4.sysctl_tcp_limit_output_bytes); limit <<= factor; if (refcount_read(&sk->sk_wmem_alloc) > limit) { -- cgit v1.2.3 From b530b68148301d73775cd27cc136ce4dd5738ae8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 27 Oct 2017 07:47:26 -0700 Subject: tcp: Namespace-ify sysctl_tcp_challenge_ack_limit Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_input.c | 14 ++++++-------- net/ipv4/tcp_ipv4.c | 2 ++ 5 files changed, 16 insertions(+), 16 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index e98f473bab13..e9895d40868e 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -147,6 +147,7 @@ struct netns_ipv4 { int sysctl_tcp_tso_win_divisor; int sysctl_tcp_workaround_signed_windows; int sysctl_tcp_limit_output_bytes; + int sysctl_tcp_challenge_ack_limit; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 33f9d30a6905..afc23596e9aa 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -250,7 +250,6 @@ extern int sysctl_tcp_rmem[3]; #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ -extern int sysctl_tcp_challenge_ack_limit; extern int sysctl_tcp_min_tso_segs; extern int sysctl_tcp_min_rtt_wlen; extern int sysctl_tcp_autocorking; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 6caf5c40730f..e28b3b7a7bbc 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -457,13 +457,6 @@ static struct ctl_table ipv4_table[] = { .maxlen = TCP_CA_NAME_MAX, .proc_handler = proc_tcp_congestion_control, }, - { - .procname = "tcp_challenge_ack_limit", - .data = &sysctl_tcp_challenge_ack_limit, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, #ifdef CONFIG_NETLABEL { .procname = "cipso_cache_enable", @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_challenge_ack_limit", + .data = &init_net.ipv4.sysctl_tcp_challenge_ack_limit, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index ce481325115f..928048a4e2c5 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -79,9 +79,6 @@ #include #include -/* rfc5961 challenge ack rate limiting */ -int sysctl_tcp_challenge_ack_limit = 1000; - int sysctl_tcp_max_orphans __read_mostly = NR_FILE; int sysctl_tcp_min_rtt_wlen __read_mostly = 300; int sysctl_tcp_invalid_ratelimit __read_mostly = HZ/2; @@ -3443,10 +3440,11 @@ static void tcp_send_challenge_ack(struct sock *sk, const struct sk_buff *skb) static u32 challenge_timestamp; static unsigned int challenge_count; struct tcp_sock *tp = tcp_sk(sk); + struct net *net = sock_net(sk); u32 count, now; /* First check our per-socket dupack rate limit. */ - if (__tcp_oow_rate_limited(sock_net(sk), + if (__tcp_oow_rate_limited(net, LINUX_MIB_TCPACKSKIPPEDCHALLENGE, &tp->last_oow_ack_time)) return; @@ -3454,16 +3452,16 @@ static void tcp_send_challenge_ack(struct sock *sk, const struct sk_buff *skb) /* Then check host-wide RFC 5961 rate limit. */ now = jiffies / HZ; if (now != challenge_timestamp) { - u32 half = (sysctl_tcp_challenge_ack_limit + 1) >> 1; + u32 ack_limit = net->ipv4.sysctl_tcp_challenge_ack_limit; + u32 half = (ack_limit + 1) >> 1; challenge_timestamp = now; - WRITE_ONCE(challenge_count, half + - prandom_u32_max(sysctl_tcp_challenge_ack_limit)); + WRITE_ONCE(challenge_count, half + prandom_u32_max(ack_limit)); } count = READ_ONCE(challenge_count); if (count > 0) { WRITE_ONCE(challenge_count, count - 1); - NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPCHALLENGEACK); + NET_INC_STATS(net, LINUX_MIB_TCPCHALLENGEACK); tcp_send_ack(sk); } } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 713b80261e4f..50ab3a3eced3 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2501,6 +2501,8 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_tso_win_divisor = 3; /* Default TSQ limit of four TSO segments */ net->ipv4.sysctl_tcp_limit_output_bytes = 262144; + /* rfc5961 challenge ack rate limiting */ + net->ipv4.sysctl_tcp_challenge_ack_limit = 1000; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); -- cgit v1.2.3 From 26e9596e5b8f11025b57b12e7265df649129ab00 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 27 Oct 2017 07:47:27 -0700 Subject: tcp: Namespace-ify sysctl_tcp_min_tso_segs Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 18 +++++++++--------- net/ipv4/tcp.c | 2 -- net/ipv4/tcp_ipv4.c | 1 + net/ipv4/tcp_output.c | 3 ++- 6 files changed, 13 insertions(+), 13 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index e9895d40868e..a2da3e19a977 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -148,6 +148,7 @@ struct netns_ipv4 { int sysctl_tcp_workaround_signed_windows; int sysctl_tcp_limit_output_bytes; int sysctl_tcp_challenge_ack_limit; + int sysctl_tcp_min_tso_segs; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index afc23596e9aa..0735303a6575 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -250,7 +250,6 @@ extern int sysctl_tcp_rmem[3]; #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ -extern int sysctl_tcp_min_tso_segs; extern int sysctl_tcp_min_rtt_wlen; extern int sysctl_tcp_autocorking; extern int sysctl_tcp_invalid_ratelimit; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index e28b3b7a7bbc..00b4aea3705b 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -499,15 +499,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_allowed_congestion_control, }, - { - .procname = "tcp_min_tso_segs", - .data = &sysctl_tcp_min_tso_segs, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &one, - .extra2 = &gso_max_segs, - }, { .procname = "tcp_pacing_ss_ratio", .data = &sysctl_tcp_pacing_ss_ratio, @@ -1145,6 +1136,15 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_min_tso_segs", + .data = &init_net.ipv4.sysctl_tcp_min_tso_segs, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &one, + .extra2 = &gso_max_segs, + }, { } }; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index c7c983f0f817..a01c97708d83 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -285,8 +285,6 @@ #include -int sysctl_tcp_min_tso_segs __read_mostly = 2; - int sysctl_tcp_autocorking __read_mostly = 1; struct percpu_counter tcp_orphan_count; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 50ab3a3eced3..6192f26145d3 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2503,6 +2503,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_limit_output_bytes = 262144; /* rfc5961 challenge ack rate limiting */ net->ipv4.sysctl_tcp_challenge_ack_limit = 1000; + net->ipv4.sysctl_tcp_min_tso_segs = 2; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index f018892c6a98..aab6e7145013 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1758,7 +1758,8 @@ static u32 tcp_tso_segs(struct sock *sk, unsigned int mss_now) u32 tso_segs = ca_ops->tso_segs_goal ? ca_ops->tso_segs_goal(sk) : 0; return tso_segs ? : - tcp_tso_autosize(sk, mss_now, sysctl_tcp_min_tso_segs); + tcp_tso_autosize(sk, mss_now, + sock_net(sk)->ipv4.sysctl_tcp_min_tso_segs); } /* Returns the portion of skb which can be sent right away */ -- cgit v1.2.3 From bd239704295c66196e6b77c5717ec4aec076ddd5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 27 Oct 2017 07:47:28 -0700 Subject: tcp: Namespace-ify sysctl_tcp_min_rtt_wlen Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_input.c | 3 +-- net/ipv4/tcp_ipv4.c | 1 + 5 files changed, 10 insertions(+), 10 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index a2da3e19a977..1a66af8a0d32 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -149,6 +149,7 @@ struct netns_ipv4 { int sysctl_tcp_limit_output_bytes; int sysctl_tcp_challenge_ack_limit; int sysctl_tcp_min_tso_segs; + int sysctl_tcp_min_rtt_wlen; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 0735303a6575..56f50c9a3e6a 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -250,7 +250,6 @@ extern int sysctl_tcp_rmem[3]; #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ -extern int sysctl_tcp_min_rtt_wlen; extern int sysctl_tcp_autocorking; extern int sysctl_tcp_invalid_ratelimit; extern int sysctl_tcp_pacing_ss_ratio; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 00b4aea3705b..029692d2e4ae 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -437,13 +437,6 @@ static struct ctl_table ipv4_table[] = { .proc_handler = proc_dointvec_minmax, .extra1 = &one, }, - { - .procname = "tcp_min_rtt_wlen", - .data = &sysctl_tcp_min_rtt_wlen, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec - }, { .procname = "tcp_low_latency", .data = &sysctl_tcp_low_latency, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .extra1 = &one, .extra2 = &gso_max_segs, }, + { + .procname = "tcp_min_rtt_wlen", + .data = &init_net.ipv4.sysctl_tcp_min_rtt_wlen, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 928048a4e2c5..da1ef666d1f9 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -80,7 +80,6 @@ #include int sysctl_tcp_max_orphans __read_mostly = NR_FILE; -int sysctl_tcp_min_rtt_wlen __read_mostly = 300; int sysctl_tcp_invalid_ratelimit __read_mostly = HZ/2; #define FLAG_DATA 0x01 /* Incoming frame contained data. */ @@ -2915,8 +2914,8 @@ static void tcp_fastretrans_alert(struct sock *sk, const int acked, static void tcp_update_rtt_min(struct sock *sk, u32 rtt_us) { + u32 wlen = sock_net(sk)->ipv4.sysctl_tcp_min_rtt_wlen * HZ; struct tcp_sock *tp = tcp_sk(sk); - u32 wlen = sysctl_tcp_min_rtt_wlen * HZ; minmax_running_min(&tp->rtt_min, wlen, tcp_jiffies32, rtt_us ? : jiffies_to_usecs(1)); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 6192f26145d3..ced35af5737a 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2504,6 +2504,7 @@ static int __net_init tcp_sk_init(struct net *net) /* rfc5961 challenge ack rate limiting */ net->ipv4.sysctl_tcp_challenge_ack_limit = 1000; net->ipv4.sysctl_tcp_min_tso_segs = 2; + net->ipv4.sysctl_tcp_min_rtt_wlen = 300; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); -- cgit v1.2.3 From 790f00e19f65673c3c169dfc137c09a9236847d5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 27 Oct 2017 07:47:29 -0700 Subject: tcp: Namespace-ify sysctl_tcp_autocorking Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 18 +++++++++--------- net/ipv4/tcp.c | 4 +--- net/ipv4/tcp_ipv4.c | 1 + 5 files changed, 12 insertions(+), 13 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 1a66af8a0d32..537830882149 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -150,6 +150,7 @@ struct netns_ipv4 { int sysctl_tcp_challenge_ack_limit; int sysctl_tcp_min_tso_segs; int sysctl_tcp_min_rtt_wlen; + int sysctl_tcp_autocorking; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 56f50c9a3e6a..0268f1025d9d 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -250,7 +250,6 @@ extern int sysctl_tcp_rmem[3]; #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ -extern int sysctl_tcp_autocorking; extern int sysctl_tcp_invalid_ratelimit; extern int sysctl_tcp_pacing_ss_ratio; extern int sysctl_tcp_pacing_ca_ratio; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 029692d2e4ae..43a18a317053 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -510,15 +510,6 @@ static struct ctl_table ipv4_table[] = { .extra1 = &zero, .extra2 = &thousand, }, - { - .procname = "tcp_autocorking", - .data = &sysctl_tcp_autocorking, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &zero, - .extra2 = &one, - }, { .procname = "tcp_invalid_ratelimit", .data = &sysctl_tcp_invalid_ratelimit, @@ -1145,6 +1136,15 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "tcp_autocorking", + .data = &init_net.ipv4.sysctl_tcp_autocorking, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &one, + }, { } }; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index a01c97708d83..a7a0f316eb86 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -285,8 +285,6 @@ #include -int sysctl_tcp_autocorking __read_mostly = 1; - struct percpu_counter tcp_orphan_count; EXPORT_SYMBOL_GPL(tcp_orphan_count); @@ -697,7 +695,7 @@ static bool tcp_should_autocork(struct sock *sk, struct sk_buff *skb, int size_goal) { return skb->len < size_goal && - sysctl_tcp_autocorking && + sock_net(sk)->ipv4.sysctl_tcp_autocorking && skb != tcp_write_queue_head(sk) && refcount_read(&sk->sk_wmem_alloc) > skb->truesize; } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index ced35af5737a..351e3497c8f3 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2505,6 +2505,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_challenge_ack_limit = 1000; net->ipv4.sysctl_tcp_min_tso_segs = 2; net->ipv4.sysctl_tcp_min_rtt_wlen = 300; + net->ipv4.sysctl_tcp_autocorking = 1; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); -- cgit v1.2.3 From 4170ba6b589ced82da56c7e4f71cc84b2be036d6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 27 Oct 2017 07:47:30 -0700 Subject: tcp: Namespace-ify sysctl_tcp_invalid_ratelimit Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 14 +++++++------- net/ipv4/tcp_input.c | 3 +-- net/ipv4/tcp_ipv4.c | 1 + 5 files changed, 10 insertions(+), 10 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 537830882149..e52c2124b32e 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -151,6 +151,7 @@ struct netns_ipv4 { int sysctl_tcp_min_tso_segs; int sysctl_tcp_min_rtt_wlen; int sysctl_tcp_autocorking; + int sysctl_tcp_invalid_ratelimit; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 0268f1025d9d..5869a822ecb1 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -250,7 +250,6 @@ extern int sysctl_tcp_rmem[3]; #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ -extern int sysctl_tcp_invalid_ratelimit; extern int sysctl_tcp_pacing_ss_ratio; extern int sysctl_tcp_pacing_ca_ratio; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 43a18a317053..6a9349c27f00 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -510,13 +510,6 @@ static struct ctl_table ipv4_table[] = { .extra1 = &zero, .extra2 = &thousand, }, - { - .procname = "tcp_invalid_ratelimit", - .data = &sysctl_tcp_invalid_ratelimit, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_ms_jiffies, - }, { .procname = "tcp_available_ulp", .maxlen = TCP_ULP_BUF_MAX, @@ -1145,6 +1138,13 @@ static struct ctl_table ipv4_net_table[] = { .extra1 = &zero, .extra2 = &one, }, + { + .procname = "tcp_invalid_ratelimit", + .data = &init_net.ipv4.sysctl_tcp_invalid_ratelimit, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_ms_jiffies, + }, { } }; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index da1ef666d1f9..db4d458d0205 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -80,7 +80,6 @@ #include int sysctl_tcp_max_orphans __read_mostly = NR_FILE; -int sysctl_tcp_invalid_ratelimit __read_mostly = HZ/2; #define FLAG_DATA 0x01 /* Incoming frame contained data. */ #define FLAG_WIN_UPDATE 0x02 /* Incoming ACK was a window update. */ @@ -3403,7 +3402,7 @@ static bool __tcp_oow_rate_limited(struct net *net, int mib_idx, if (*last_oow_ack_time) { s32 elapsed = (s32)(tcp_jiffies32 - *last_oow_ack_time); - if (0 <= elapsed && elapsed < sysctl_tcp_invalid_ratelimit) { + if (0 <= elapsed && elapsed < net->ipv4.sysctl_tcp_invalid_ratelimit) { NET_INC_STATS(net, mib_idx); return true; /* rate-limited: don't send yet! */ } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 351e3497c8f3..6617aae18ba2 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2506,6 +2506,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_min_tso_segs = 2; net->ipv4.sysctl_tcp_min_rtt_wlen = 300; net->ipv4.sysctl_tcp_autocorking = 1; + net->ipv4.sysctl_tcp_invalid_ratelimit = HZ/2; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); -- cgit v1.2.3 From 23a7102a2d1068508fa2a0ce593a0df7f8fdc0ac Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 27 Oct 2017 07:47:31 -0700 Subject: tcp: Namespace-ify sysctl_tcp_pacing_ss_ratio Also remove an obsolete comment about TCP pacing. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 1 - net/ipv4/sysctl_net_ipv4.c | 18 +++++++++--------- net/ipv4/tcp_input.c | 9 +-------- net/ipv4/tcp_ipv4.c | 1 + 5 files changed, 12 insertions(+), 18 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index e52c2124b32e..eb2dcf1cbe61 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -152,6 +152,7 @@ struct netns_ipv4 { int sysctl_tcp_min_rtt_wlen; int sysctl_tcp_autocorking; int sysctl_tcp_invalid_ratelimit; + int sysctl_tcp_pacing_ss_ratio; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 5869a822ecb1..2a5f8261ca03 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -250,7 +250,6 @@ extern int sysctl_tcp_rmem[3]; #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ -extern int sysctl_tcp_pacing_ss_ratio; extern int sysctl_tcp_pacing_ca_ratio; extern atomic_long_t tcp_memory_allocated; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 6a9349c27f00..7f0dba852d47 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -492,15 +492,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_allowed_congestion_control, }, - { - .procname = "tcp_pacing_ss_ratio", - .data = &sysctl_tcp_pacing_ss_ratio, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &zero, - .extra2 = &thousand, - }, { .procname = "tcp_pacing_ca_ratio", .data = &sysctl_tcp_pacing_ca_ratio, @@ -1145,6 +1136,15 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = proc_dointvec_ms_jiffies, }, + { + .procname = "tcp_pacing_ss_ratio", + .data = &init_net.ipv4.sysctl_tcp_pacing_ss_ratio, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &thousand, + }, { } }; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index db4d458d0205..29539d39e61a 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -767,13 +767,6 @@ static void tcp_rtt_estimator(struct sock *sk, long mrtt_us) tp->srtt_us = max(1U, srtt); } -/* Set the sk_pacing_rate to allow proper sizing of TSO packets. - * Note: TCP stack does not yet implement pacing. - * FQ packet scheduler can be used to implement cheap but effective - * TCP pacing, to smooth the burst on large writes when packets - * in flight is significantly lower than cwnd (or rwin) - */ -int sysctl_tcp_pacing_ss_ratio __read_mostly = 200; int sysctl_tcp_pacing_ca_ratio __read_mostly = 120; static void tcp_update_pacing_rate(struct sock *sk) @@ -793,7 +786,7 @@ static void tcp_update_pacing_rate(struct sock *sk) * end of slow start and should slow down. */ if (tp->snd_cwnd < tp->snd_ssthresh / 2) - rate *= sysctl_tcp_pacing_ss_ratio; + rate *= sock_net(sk)->ipv4.sysctl_tcp_pacing_ss_ratio; else rate *= sysctl_tcp_pacing_ca_ratio; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 6617aae18ba2..1d8fc663af51 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2507,6 +2507,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_min_rtt_wlen = 300; net->ipv4.sysctl_tcp_autocorking = 1; net->ipv4.sysctl_tcp_invalid_ratelimit = HZ/2; + net->ipv4.sysctl_tcp_pacing_ss_ratio = 200; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); -- cgit v1.2.3 From c26e91f8b9b8e1fd252e07c1f60e50220cd7ebab Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 27 Oct 2017 07:47:32 -0700 Subject: tcp: Namespace-ify sysctl_tcp_pacing_ca_ratio Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 2 -- net/ipv4/sysctl_net_ipv4.c | 18 +++++++++--------- net/ipv4/tcp_input.c | 4 +--- net/ipv4/tcp_ipv4.c | 1 + 5 files changed, 12 insertions(+), 14 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index eb2dcf1cbe61..141ba82b5efb 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -153,6 +153,7 @@ struct netns_ipv4 { int sysctl_tcp_autocorking; int sysctl_tcp_invalid_ratelimit; int sysctl_tcp_pacing_ss_ratio; + int sysctl_tcp_pacing_ca_ratio; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index 2a5f8261ca03..092d606fcc16 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -250,8 +250,6 @@ extern int sysctl_tcp_rmem[3]; #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ -extern int sysctl_tcp_pacing_ca_ratio; - extern atomic_long_t tcp_memory_allocated; extern struct percpu_counter tcp_sockets_allocated; extern unsigned long tcp_memory_pressure; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 7f0dba852d47..4602af6d5358 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -492,15 +492,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_allowed_congestion_control, }, - { - .procname = "tcp_pacing_ca_ratio", - .data = &sysctl_tcp_pacing_ca_ratio, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &zero, - .extra2 = &thousand, - }, { .procname = "tcp_available_ulp", .maxlen = TCP_ULP_BUF_MAX, @@ -1145,6 +1136,15 @@ static struct ctl_table ipv4_net_table[] = { .extra1 = &zero, .extra2 = &thousand, }, + { + .procname = "tcp_pacing_ca_ratio", + .data = &init_net.ipv4.sysctl_tcp_pacing_ca_ratio, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &thousand, + }, { } }; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 29539d39e61a..21c358c0cf2e 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -767,8 +767,6 @@ static void tcp_rtt_estimator(struct sock *sk, long mrtt_us) tp->srtt_us = max(1U, srtt); } -int sysctl_tcp_pacing_ca_ratio __read_mostly = 120; - static void tcp_update_pacing_rate(struct sock *sk) { const struct tcp_sock *tp = tcp_sk(sk); @@ -788,7 +786,7 @@ static void tcp_update_pacing_rate(struct sock *sk) if (tp->snd_cwnd < tp->snd_ssthresh / 2) rate *= sock_net(sk)->ipv4.sysctl_tcp_pacing_ss_ratio; else - rate *= sysctl_tcp_pacing_ca_ratio; + rate *= sock_net(sk)->ipv4.sysctl_tcp_pacing_ca_ratio; rate *= max(tp->snd_cwnd, tp->packets_out); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 1d8fc663af51..7c1dae6493c3 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2508,6 +2508,7 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_autocorking = 1; net->ipv4.sysctl_tcp_invalid_ratelimit = HZ/2; net->ipv4.sysctl_tcp_pacing_ss_ratio = 200; + net->ipv4.sysctl_tcp_pacing_ca_ratio = 120; net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); -- cgit v1.2.3 From 5b52a4c3acf5f4b4854d1c3ddc8be8770330a79c Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Fri, 27 Oct 2017 10:01:39 -0700 Subject: tcp: remove unnecessary include two extra #include are not necessary in tcp.h Remove them. Fixes: 40304b2a1567 ("bpf: BPF support for sock_ops") Signed-off-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/net/tcp.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'include/net') diff --git a/include/net/tcp.h b/include/net/tcp.h index 092d606fcc16..aa1cc90fdc02 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -45,9 +45,6 @@ #include #include - -#include -#include #include extern struct inet_hashinfo tcp_hashinfo; -- cgit v1.2.3 From 1f01d8be0e6a04bd682a55f6d50c14c1679e7571 Mon Sep 17 00:00:00 2001 From: Konrad Zapałowicz Date: Tue, 17 Oct 2017 15:53:49 +0200 Subject: Bluetooth: increase timeout for le auto connections MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch increases the connection timeout for LE connections that are triggered by the advertising report to 4 seconds. It has been observed that devices equipped with wifi+bt combo SoC fail to create a connection with BLE devices due to their coexistence issues. Increasing this timeout gives them enough time to complete the connection with success. Signed-off-by: Konrad Zapałowicz Signed-off-by: Marcel Holtmann --- include/net/bluetooth/hci.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index fe98f0a5bef0..1668211297a9 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -273,7 +273,7 @@ enum { #define HCI_AUTO_OFF_TIMEOUT msecs_to_jiffies(2000) /* 2 seconds */ #define HCI_POWER_OFF_TIMEOUT msecs_to_jiffies(5000) /* 5 seconds */ #define HCI_LE_CONN_TIMEOUT msecs_to_jiffies(20000) /* 20 seconds */ -#define HCI_LE_AUTOCONN_TIMEOUT msecs_to_jiffies(2000) /* 2 seconds */ +#define HCI_LE_AUTOCONN_TIMEOUT msecs_to_jiffies(4000) /* 4 seconds */ /* HCI data types */ #define HCI_COMMAND_PKT 0x01 -- cgit v1.2.3 From 2064ee332e4c1b7495cf68b84355c213d8fe71fd Mon Sep 17 00:00:00 2001 From: Marcel Holtmann Date: Mon, 30 Oct 2017 10:42:59 +0100 Subject: Bluetooth: Use bt_dev_err and bt_dev_info when possible In case of using BT_ERR and BT_INFO, convert to bt_dev_err and bt_dev_info when possible. This allows for controller specific reporting. Signed-off-by: Marcel Holtmann Signed-off-by: Johan Hedberg --- drivers/bluetooth/bpa10x.c | 15 ++-- drivers/bluetooth/bt3c_cs.c | 2 +- drivers/bluetooth/btbcm.c | 103 ++++++++++++------------- drivers/bluetooth/btintel.c | 42 +++++------ drivers/bluetooth/btqca.c | 6 +- drivers/bluetooth/btrtl.c | 21 +++--- drivers/bluetooth/btusb.c | 155 ++++++++++++++++++-------------------- drivers/bluetooth/hci_ath.c | 2 +- drivers/bluetooth/hci_h4.c | 2 +- drivers/bluetooth/hci_qca.c | 12 +-- drivers/bluetooth/hci_serdev.c | 9 +-- include/net/bluetooth/bluetooth.h | 3 + net/bluetooth/a2mp.c | 2 +- net/bluetooth/amp.c | 4 +- net/bluetooth/hci_conn.c | 4 +- net/bluetooth/hci_core.c | 35 +++++---- net/bluetooth/hci_event.c | 46 +++++------ net/bluetooth/hci_request.c | 16 ++-- net/bluetooth/hci_sysfs.c | 2 +- net/bluetooth/mgmt.c | 56 +++++++------- net/bluetooth/smp.c | 22 +++--- 21 files changed, 279 insertions(+), 280 deletions(-) (limited to 'include/net') diff --git a/drivers/bluetooth/bpa10x.c b/drivers/bluetooth/bpa10x.c index 48d10cb5c9a1..7971bfbd4321 100644 --- a/drivers/bluetooth/bpa10x.c +++ b/drivers/bluetooth/bpa10x.c @@ -117,7 +117,7 @@ static void bpa10x_rx_complete(struct urb *urb) bpa10x_recv_pkts, ARRAY_SIZE(bpa10x_recv_pkts)); if (IS_ERR(data->rx_skb[idx])) { - BT_ERR("%s corrupted event packet", hdev->name); + bt_dev_err(hdev, "corrupted event packet"); hdev->stat.err_rx++; data->rx_skb[idx] = NULL; } @@ -127,8 +127,7 @@ static void bpa10x_rx_complete(struct urb *urb) err = usb_submit_urb(urb, GFP_ATOMIC); if (err < 0) { - BT_ERR("%s urb %p failed to resubmit (%d)", - hdev->name, urb, -err); + bt_dev_err(hdev, "urb %p failed to resubmit (%d)", urb, -err); usb_unanchor_urb(urb); } } @@ -164,8 +163,7 @@ static inline int bpa10x_submit_intr_urb(struct hci_dev *hdev) err = usb_submit_urb(urb, GFP_KERNEL); if (err < 0) { - BT_ERR("%s urb %p submission failed (%d)", - hdev->name, urb, -err); + bt_dev_err(hdev, "urb %p submission failed (%d)", urb, -err); usb_unanchor_urb(urb); } @@ -205,8 +203,7 @@ static inline int bpa10x_submit_bulk_urb(struct hci_dev *hdev) err = usb_submit_urb(urb, GFP_KERNEL); if (err < 0) { - BT_ERR("%s urb %p submission failed (%d)", - hdev->name, urb, -err); + bt_dev_err(hdev, "urb %p submission failed (%d)", urb, -err); usb_unanchor_urb(urb); } @@ -272,7 +269,7 @@ static int bpa10x_setup(struct hci_dev *hdev) if (IS_ERR(skb)) return PTR_ERR(skb); - BT_INFO("%s: %s", hdev->name, (char *)(skb->data + 1)); + bt_dev_info(hdev, "%s", (char *)(skb->data + 1)); hci_set_fw_info(hdev, "%s", skb->data + 1); @@ -348,7 +345,7 @@ static int bpa10x_send_frame(struct hci_dev *hdev, struct sk_buff *skb) err = usb_submit_urb(urb, GFP_ATOMIC); if (err < 0) { - BT_ERR("%s urb %p submission failed", hdev->name, urb); + bt_dev_err(hdev, "urb %p submission failed", urb); kfree(urb->setup_packet); usb_unanchor_urb(urb); } diff --git a/drivers/bluetooth/bt3c_cs.c b/drivers/bluetooth/bt3c_cs.c index 194788739a83..25b0cf952b91 100644 --- a/drivers/bluetooth/bt3c_cs.c +++ b/drivers/bluetooth/bt3c_cs.c @@ -355,7 +355,7 @@ static irqreturn_t bt3c_interrupt(int irq, void *dev_inst) } else if ((stat & 0xff) != 0xff) { if (stat & 0x0020) { int status = bt3c_read(iobase, 0x7002) & 0x10; - BT_INFO("%s: Antenna %s", info->hdev->name, + bt_dev_info(info->hdev, "Antenna %s", status ? "out" : "in"); } if (stat & 0x0001) diff --git a/drivers/bluetooth/btbcm.c b/drivers/bluetooth/btbcm.c index ae1fa390f508..afa4cb3b16e3 100644 --- a/drivers/bluetooth/btbcm.c +++ b/drivers/bluetooth/btbcm.c @@ -45,13 +45,12 @@ int btbcm_check_bdaddr(struct hci_dev *hdev) HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { int err = PTR_ERR(skb); - BT_ERR("%s: BCM: Reading device address failed (%d)", - hdev->name, err); + bt_dev_err(hdev, "BCM: Reading device address failed (%d)", err); return err; } if (skb->len != sizeof(*bda)) { - BT_ERR("%s: BCM: Device address length mismatch", hdev->name); + bt_dev_err(hdev, "BCM: Device address length mismatch"); kfree_skb(skb); return -EIO; } @@ -74,8 +73,8 @@ int btbcm_check_bdaddr(struct hci_dev *hdev) if (!bacmp(&bda->bdaddr, BDADDR_BCM20702A0) || !bacmp(&bda->bdaddr, BDADDR_BCM4324B3) || !bacmp(&bda->bdaddr, BDADDR_BCM4330B1)) { - BT_INFO("%s: BCM: Using default device address (%pMR)", - hdev->name, &bda->bdaddr); + bt_dev_info(hdev, "BCM: Using default device address (%pMR)", + &bda->bdaddr); set_bit(HCI_QUIRK_INVALID_BDADDR, &hdev->quirks); } @@ -93,8 +92,7 @@ int btbcm_set_bdaddr(struct hci_dev *hdev, const bdaddr_t *bdaddr) skb = __hci_cmd_sync(hdev, 0xfc01, 6, bdaddr, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { err = PTR_ERR(skb); - BT_ERR("%s: BCM: Change address command failed (%d)", - hdev->name, err); + bt_dev_err(hdev, "BCM: Change address command failed (%d)", err); return err; } kfree_skb(skb); @@ -116,8 +114,8 @@ int btbcm_patchram(struct hci_dev *hdev, const struct firmware *fw) skb = __hci_cmd_sync(hdev, 0xfc2e, 0, NULL, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { err = PTR_ERR(skb); - BT_ERR("%s: BCM: Download Minidrv command failed (%d)", - hdev->name, err); + bt_dev_err(hdev, "BCM: Download Minidrv command failed (%d)", + err); goto done; } kfree_skb(skb); @@ -136,7 +134,7 @@ int btbcm_patchram(struct hci_dev *hdev, const struct firmware *fw) fw_size -= sizeof(*cmd); if (fw_size < cmd->plen) { - BT_ERR("%s: BCM: Patch is corrupted", hdev->name); + bt_dev_err(hdev, "BCM: Patch is corrupted"); err = -EINVAL; goto done; } @@ -151,8 +149,8 @@ int btbcm_patchram(struct hci_dev *hdev, const struct firmware *fw) HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { err = PTR_ERR(skb); - BT_ERR("%s: BCM: Patch command %04x failed (%d)", - hdev->name, opcode, err); + bt_dev_err(hdev, "BCM: Patch command %04x failed (%d)", + opcode, err); goto done; } kfree_skb(skb); @@ -173,7 +171,7 @@ static int btbcm_reset(struct hci_dev *hdev) skb = __hci_cmd_sync(hdev, HCI_OP_RESET, 0, NULL, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { int err = PTR_ERR(skb); - BT_ERR("%s: BCM: Reset failed (%d)", hdev->name, err); + bt_dev_err(hdev, "BCM: Reset failed (%d)", err); return err; } kfree_skb(skb); @@ -191,13 +189,13 @@ static struct sk_buff *btbcm_read_local_name(struct hci_dev *hdev) skb = __hci_cmd_sync(hdev, HCI_OP_READ_LOCAL_NAME, 0, NULL, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { - BT_ERR("%s: BCM: Reading local name failed (%ld)", - hdev->name, PTR_ERR(skb)); + bt_dev_err(hdev, "BCM: Reading local name failed (%ld)", + PTR_ERR(skb)); return skb; } if (skb->len != sizeof(struct hci_rp_read_local_name)) { - BT_ERR("%s: BCM: Local name length mismatch", hdev->name); + bt_dev_err(hdev, "BCM: Local name length mismatch"); kfree_skb(skb); return ERR_PTR(-EIO); } @@ -212,13 +210,13 @@ static struct sk_buff *btbcm_read_local_version(struct hci_dev *hdev) skb = __hci_cmd_sync(hdev, HCI_OP_READ_LOCAL_VERSION, 0, NULL, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { - BT_ERR("%s: BCM: Reading local version info failed (%ld)", - hdev->name, PTR_ERR(skb)); + bt_dev_err(hdev, "BCM: Reading local version info failed (%ld)", + PTR_ERR(skb)); return skb; } if (skb->len != sizeof(struct hci_rp_read_local_version)) { - BT_ERR("%s: BCM: Local version length mismatch", hdev->name); + bt_dev_err(hdev, "BCM: Local version length mismatch"); kfree_skb(skb); return ERR_PTR(-EIO); } @@ -232,13 +230,13 @@ static struct sk_buff *btbcm_read_verbose_config(struct hci_dev *hdev) skb = __hci_cmd_sync(hdev, 0xfc79, 0, NULL, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { - BT_ERR("%s: BCM: Read verbose config info failed (%ld)", - hdev->name, PTR_ERR(skb)); + bt_dev_err(hdev, "BCM: Read verbose config info failed (%ld)", + PTR_ERR(skb)); return skb; } if (skb->len != 7) { - BT_ERR("%s: BCM: Verbose config length mismatch", hdev->name); + bt_dev_err(hdev, "BCM: Verbose config length mismatch"); kfree_skb(skb); return ERR_PTR(-EIO); } @@ -252,14 +250,13 @@ static struct sk_buff *btbcm_read_controller_features(struct hci_dev *hdev) skb = __hci_cmd_sync(hdev, 0xfc6e, 0, NULL, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { - BT_ERR("%s: BCM: Read controller features failed (%ld)", - hdev->name, PTR_ERR(skb)); + bt_dev_err(hdev, "BCM: Read controller features failed (%ld)", + PTR_ERR(skb)); return skb; } if (skb->len != 9) { - BT_ERR("%s: BCM: Controller features length mismatch", - hdev->name); + bt_dev_err(hdev, "BCM: Controller features length mismatch"); kfree_skb(skb); return ERR_PTR(-EIO); } @@ -273,13 +270,13 @@ static struct sk_buff *btbcm_read_usb_product(struct hci_dev *hdev) skb = __hci_cmd_sync(hdev, 0xfc5a, 0, NULL, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { - BT_ERR("%s: BCM: Read USB product info failed (%ld)", - hdev->name, PTR_ERR(skb)); + bt_dev_err(hdev, "BCM: Read USB product info failed (%ld)", + PTR_ERR(skb)); return skb; } if (skb->len != 5) { - BT_ERR("%s: BCM: USB product length mismatch", hdev->name); + bt_dev_err(hdev, "BCM: USB product length mismatch"); kfree_skb(skb); return ERR_PTR(-EIO); } @@ -296,7 +293,7 @@ static int btbcm_read_info(struct hci_dev *hdev) if (IS_ERR(skb)) return PTR_ERR(skb); - BT_INFO("%s: BCM: chip id %u", hdev->name, skb->data[1]); + bt_dev_info(hdev, "BCM: chip id %u", skb->data[1]); kfree_skb(skb); /* Read Controller Features */ @@ -304,7 +301,7 @@ static int btbcm_read_info(struct hci_dev *hdev) if (IS_ERR(skb)) return PTR_ERR(skb); - BT_INFO("%s: BCM: features 0x%2.2x", hdev->name, skb->data[1]); + bt_dev_info(hdev, "BCM: features 0x%2.2x", skb->data[1]); kfree_skb(skb); /* Read Local Name */ @@ -312,7 +309,7 @@ static int btbcm_read_info(struct hci_dev *hdev) if (IS_ERR(skb)) return PTR_ERR(skb); - BT_INFO("%s: %s", hdev->name, (char *)(skb->data + 1)); + bt_dev_info(hdev, "%s", (char *)(skb->data + 1)); kfree_skb(skb); return 0; @@ -378,9 +375,9 @@ int btbcm_initialize(struct hci_dev *hdev, char *fw_name, size_t len) return 0; } - BT_INFO("%s: %s (%3.3u.%3.3u.%3.3u) build %4.4u", hdev->name, - hw_name ? : "BCM", (subver & 0xe000) >> 13, - (subver & 0x1f00) >> 8, (subver & 0x00ff), rev & 0x0fff); + bt_dev_info(hdev, "%s (%3.3u.%3.3u.%3.3u) build %4.4u", + hw_name ? : "BCM", (subver & 0xe000) >> 13, + (subver & 0x1f00) >> 8, (subver & 0x00ff), rev & 0x0fff); return 0; } @@ -408,9 +405,9 @@ int btbcm_finalize(struct hci_dev *hdev) subver = le16_to_cpu(ver->lmp_subver); kfree_skb(skb); - BT_INFO("%s: BCM (%3.3u.%3.3u.%3.3u) build %4.4u", hdev->name, - (subver & 0xe000) >> 13, (subver & 0x1f00) >> 8, - (subver & 0x00ff), rev & 0x0fff); + bt_dev_info(hdev, "BCM (%3.3u.%3.3u.%3.3u) build %4.4u", + (subver & 0xe000) >> 13, (subver & 0x1f00) >> 8, + (subver & 0x00ff), rev & 0x0fff); btbcm_check_bdaddr(hdev); @@ -505,13 +502,13 @@ int btbcm_setup_patchram(struct hci_dev *hdev) return 0; } - BT_INFO("%s: %s (%3.3u.%3.3u.%3.3u) build %4.4u", hdev->name, - hw_name ? : "BCM", (subver & 0xe000) >> 13, - (subver & 0x1f00) >> 8, (subver & 0x00ff), rev & 0x0fff); + bt_dev_info(hdev, "%s (%3.3u.%3.3u.%3.3u) build %4.4u", + hw_name ? : "BCM", (subver & 0xe000) >> 13, + (subver & 0x1f00) >> 8, (subver & 0x00ff), rev & 0x0fff); err = request_firmware(&fw, fw_name, &hdev->dev); if (err < 0) { - BT_INFO("%s: BCM: Patch %s not found", hdev->name, fw_name); + bt_dev_info(hdev, "BCM: Patch %s not found", fw_name); goto done; } @@ -534,16 +531,16 @@ int btbcm_setup_patchram(struct hci_dev *hdev) subver = le16_to_cpu(ver->lmp_subver); kfree_skb(skb); - BT_INFO("%s: %s (%3.3u.%3.3u.%3.3u) build %4.4u", hdev->name, - hw_name ? : "BCM", (subver & 0xe000) >> 13, - (subver & 0x1f00) >> 8, (subver & 0x00ff), rev & 0x0fff); + bt_dev_info(hdev, "%s (%3.3u.%3.3u.%3.3u) build %4.4u", + hw_name ? : "BCM", (subver & 0xe000) >> 13, + (subver & 0x1f00) >> 8, (subver & 0x00ff), rev & 0x0fff); /* Read Local Name */ skb = btbcm_read_local_name(hdev); if (IS_ERR(skb)) return PTR_ERR(skb); - BT_INFO("%s: %s", hdev->name, (char *)(skb->data + 1)); + bt_dev_info(hdev, "%s", (char *)(skb->data + 1)); kfree_skb(skb); done: @@ -568,31 +565,31 @@ int btbcm_setup_apple(struct hci_dev *hdev) /* Read Verbose Config Version Info */ skb = btbcm_read_verbose_config(hdev); if (!IS_ERR(skb)) { - BT_INFO("%s: BCM: chip id %u build %4.4u", hdev->name, - skb->data[1], get_unaligned_le16(skb->data + 5)); + bt_dev_info(hdev, "BCM: chip id %u build %4.4u", + skb->data[1], get_unaligned_le16(skb->data + 5)); kfree_skb(skb); } /* Read USB Product Info */ skb = btbcm_read_usb_product(hdev); if (!IS_ERR(skb)) { - BT_INFO("%s: BCM: product %4.4x:%4.4x", hdev->name, - get_unaligned_le16(skb->data + 1), - get_unaligned_le16(skb->data + 3)); + bt_dev_info(hdev, "BCM: product %4.4x:%4.4x", + get_unaligned_le16(skb->data + 1), + get_unaligned_le16(skb->data + 3)); kfree_skb(skb); } /* Read Controller Features */ skb = btbcm_read_controller_features(hdev); if (!IS_ERR(skb)) { - BT_INFO("%s: BCM: features 0x%2.2x", hdev->name, skb->data[1]); + bt_dev_info(hdev, "BCM: features 0x%2.2x", skb->data[1]); kfree_skb(skb); } /* Read Local Name */ skb = btbcm_read_local_name(hdev); if (!IS_ERR(skb)) { - BT_INFO("%s: %s", hdev->name, (char *)(skb->data + 1)); + bt_dev_info(hdev, "%s", (char *)(skb->data + 1)); kfree_skb(skb); } diff --git a/drivers/bluetooth/btintel.c b/drivers/bluetooth/btintel.c index d32e109bd5cb..4459555c9d88 100644 --- a/drivers/bluetooth/btintel.c +++ b/drivers/bluetooth/btintel.c @@ -43,13 +43,13 @@ int btintel_check_bdaddr(struct hci_dev *hdev) HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { int err = PTR_ERR(skb); - BT_ERR("%s: Reading Intel device address failed (%d)", - hdev->name, err); + bt_dev_err(hdev, "Reading Intel device address failed (%d)", + err); return err; } if (skb->len != sizeof(*bda)) { - BT_ERR("%s: Intel device address length mismatch", hdev->name); + bt_dev_err(hdev, "Intel device address length mismatch"); kfree_skb(skb); return -EIO; } @@ -62,8 +62,8 @@ int btintel_check_bdaddr(struct hci_dev *hdev) * and that in turn can cause problems with Bluetooth operation. */ if (!bacmp(&bda->bdaddr, BDADDR_INTEL)) { - BT_ERR("%s: Found Intel default device address (%pMR)", - hdev->name, &bda->bdaddr); + bt_dev_err(hdev, "Found Intel default device address (%pMR)", + &bda->bdaddr); set_bit(HCI_QUIRK_INVALID_BDADDR, &hdev->quirks); } @@ -123,8 +123,8 @@ int btintel_set_bdaddr(struct hci_dev *hdev, const bdaddr_t *bdaddr) skb = __hci_cmd_sync(hdev, 0xfc31, 6, bdaddr, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { err = PTR_ERR(skb); - BT_ERR("%s: Changing Intel device address failed (%d)", - hdev->name, err); + bt_dev_err(hdev, "Changing Intel device address failed (%d)", + err); return err; } kfree_skb(skb); @@ -154,8 +154,8 @@ int btintel_set_diag(struct hci_dev *hdev, bool enable) err = PTR_ERR(skb); if (err == -ENODATA) goto done; - BT_ERR("%s: Changing Intel diagnostic mode failed (%d)", - hdev->name, err); + bt_dev_err(hdev, "Changing Intel diagnostic mode failed (%d)", + err); return err; } kfree_skb(skb); @@ -189,30 +189,30 @@ void btintel_hw_error(struct hci_dev *hdev, u8 code) struct sk_buff *skb; u8 type = 0x00; - BT_ERR("%s: Hardware error 0x%2.2x", hdev->name, code); + bt_dev_err(hdev, "Hardware error 0x%2.2x", code); skb = __hci_cmd_sync(hdev, HCI_OP_RESET, 0, NULL, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { - BT_ERR("%s: Reset after hardware error failed (%ld)", - hdev->name, PTR_ERR(skb)); + bt_dev_err(hdev, "Reset after hardware error failed (%ld)", + PTR_ERR(skb)); return; } kfree_skb(skb); skb = __hci_cmd_sync(hdev, 0xfc22, 1, &type, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { - BT_ERR("%s: Retrieving Intel exception info failed (%ld)", - hdev->name, PTR_ERR(skb)); + bt_dev_err(hdev, "Retrieving Intel exception info failed (%ld)", + PTR_ERR(skb)); return; } if (skb->len != 13) { - BT_ERR("%s: Exception info size mismatch", hdev->name); + bt_dev_err(hdev, "Exception info size mismatch"); kfree_skb(skb); return; } - BT_ERR("%s: Exception info %s", hdev->name, (char *)(skb->data + 1)); + bt_dev_err(hdev, "Exception info %s", (char *)(skb->data + 1)); kfree_skb(skb); } @@ -233,9 +233,10 @@ void btintel_version_info(struct hci_dev *hdev, struct intel_version *ver) return; } - BT_INFO("%s: %s revision %u.%u build %u week %u %u", hdev->name, - variant, ver->fw_revision >> 4, ver->fw_revision & 0x0f, - ver->fw_build_num, ver->fw_build_ww, 2000 + ver->fw_build_yy); + bt_dev_info(hdev, "%s revision %u.%u build %u week %u %u", + variant, ver->fw_revision >> 4, ver->fw_revision & 0x0f, + ver->fw_build_num, ver->fw_build_ww, + 2000 + ver->fw_build_yy); } EXPORT_SYMBOL_GPL(btintel_version_info); @@ -321,8 +322,7 @@ int btintel_set_event_mask(struct hci_dev *hdev, bool debug) skb = __hci_cmd_sync(hdev, 0xfc52, 8, mask, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { err = PTR_ERR(skb); - BT_ERR("%s: Setting Intel event mask failed (%d)", - hdev->name, err); + bt_dev_err(hdev, "Setting Intel event mask failed (%d)", err); return err; } kfree_skb(skb); diff --git a/drivers/bluetooth/btqca.c b/drivers/bluetooth/btqca.c index 0bbdfcef2aa8..2793d4180d2f 100644 --- a/drivers/bluetooth/btqca.c +++ b/drivers/bluetooth/btqca.c @@ -287,7 +287,7 @@ static int rome_download_firmware(struct hci_dev *hdev, const struct firmware *fw; int ret; - BT_INFO("%s: ROME Downloading %s", hdev->name, config->fwname); + bt_dev_info(hdev, "ROME Downloading %s", config->fwname); ret = request_firmware(&fw, config->fwname, &hdev->dev); if (ret) { @@ -351,7 +351,7 @@ int qca_uart_setup_rome(struct hci_dev *hdev, uint8_t baudrate) return err; } - BT_INFO("%s: ROME controller version 0x%08x", hdev->name, rome_ver); + bt_dev_info(hdev, "ROME controller version 0x%08x", rome_ver); /* Download rampatch file */ config.type = TLV_TYPE_PATCH; @@ -380,7 +380,7 @@ int qca_uart_setup_rome(struct hci_dev *hdev, uint8_t baudrate) return err; } - BT_INFO("%s: ROME setup on UART is completed", hdev->name); + bt_dev_info(hdev, "ROME setup on UART is completed"); return 0; } diff --git a/drivers/bluetooth/btrtl.c b/drivers/bluetooth/btrtl.c index d9a99b4302ea..6e2ad748abba 100644 --- a/drivers/bluetooth/btrtl.c +++ b/drivers/bluetooth/btrtl.c @@ -55,8 +55,8 @@ static int rtl_read_rom_version(struct hci_dev *hdev, u8 *version) } rom_version = (struct rtl_rom_version_evt *)skb->data; - BT_INFO("%s: rom_version status=%x version=%x", - hdev->name, rom_version->status, rom_version->version); + bt_dev_info(hdev, "rom_version status=%x version=%x", + rom_version->status, rom_version->version); *version = rom_version->version; @@ -273,7 +273,7 @@ static int rtl_load_config(struct hci_dev *hdev, const char *name, u8 **buff) const struct firmware *fw; int ret; - BT_INFO("%s: rtl: loading %s", hdev->name, name); + bt_dev_info(hdev, "rtl: loading %s", name); ret = request_firmware(&fw, name, &hdev->dev); if (ret < 0) return ret; @@ -292,7 +292,7 @@ static int btrtl_setup_rtl8723a(struct hci_dev *hdev) const struct firmware *fw; int ret; - BT_INFO("%s: rtl: loading rtl_bt/rtl8723a_fw.bin", hdev->name); + bt_dev_info(hdev, "rtl: loading rtl_bt/rtl8723a_fw.bin"); ret = request_firmware(&fw, "rtl_bt/rtl8723a_fw.bin", &hdev->dev); if (ret < 0) { BT_ERR("%s: Failed to load rtl_bt/rtl8723a_fw.bin", hdev->name); @@ -363,7 +363,7 @@ static int btrtl_setup_rtl8723b(struct hci_dev *hdev, u16 lmp_subver, } else cfg_sz = 0; - BT_INFO("%s: rtl: loading %s", hdev->name, fw_name); + bt_dev_info(hdev, "rtl: loading %s", fw_name); ret = request_firmware(&fw, fw_name, &hdev->dev); if (ret < 0) { BT_ERR("%s: Failed to load %s", hdev->name, fw_name); @@ -390,7 +390,7 @@ static int btrtl_setup_rtl8723b(struct hci_dev *hdev, u16 lmp_subver, fw_data = tbuff; } - BT_INFO("cfg_sz %d, total size %d", cfg_sz, ret); + bt_dev_info(hdev, "cfg_sz %d, total size %d", cfg_sz, ret); ret = rtl_download_firmware(hdev, fw_data, ret); @@ -436,9 +436,10 @@ int btrtl_setup_realtek(struct hci_dev *hdev) return -PTR_ERR(skb); resp = (struct hci_rp_read_local_version *)skb->data; - BT_INFO("%s: rtl: examining hci_ver=%02x hci_rev=%04x lmp_ver=%02x " - "lmp_subver=%04x", hdev->name, resp->hci_ver, resp->hci_rev, - resp->lmp_ver, resp->lmp_subver); + bt_dev_info(hdev, "rtl: examining hci_ver=%02x hci_rev=%04x " + "lmp_ver=%02x lmp_subver=%04x", + resp->hci_ver, resp->hci_rev, + resp->lmp_ver, resp->lmp_subver); lmp_subver = le16_to_cpu(resp->lmp_subver); kfree_skb(skb); @@ -466,7 +467,7 @@ int btrtl_setup_realtek(struct hci_dev *hdev) return btrtl_setup_rtl8723b(hdev, lmp_subver, "rtl_bt/rtl8822b_fw.bin"); default: - BT_INFO("rtl: assuming no firmware upload needed."); + bt_dev_info(hdev, "rtl: assuming no firmware upload needed"); return 0; } } diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index b8eb39436ef2..8701140e11f9 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -643,7 +643,7 @@ static void btusb_intr_complete(struct urb *urb) if (btusb_recv_intr(data, urb->transfer_buffer, urb->actual_length) < 0) { - BT_ERR("%s corrupted event packet", hdev->name); + bt_dev_err(hdev, "corrupted event packet"); hdev->stat.err_rx++; } } else if (urb->status == -ENOENT) { @@ -663,8 +663,8 @@ static void btusb_intr_complete(struct urb *urb) * -ENODEV: device got disconnected */ if (err != -EPERM && err != -ENODEV) - BT_ERR("%s urb %p failed to resubmit (%d)", - hdev->name, urb, -err); + bt_dev_err(hdev, "urb %p failed to resubmit (%d)", + urb, -err); usb_unanchor_urb(urb); } } @@ -706,8 +706,8 @@ static int btusb_submit_intr_urb(struct hci_dev *hdev, gfp_t mem_flags) err = usb_submit_urb(urb, mem_flags); if (err < 0) { if (err != -EPERM && err != -ENODEV) - BT_ERR("%s urb %p submission failed (%d)", - hdev->name, urb, -err); + bt_dev_err(hdev, "urb %p submission failed (%d)", + urb, -err); usb_unanchor_urb(urb); } @@ -733,7 +733,7 @@ static void btusb_bulk_complete(struct urb *urb) if (data->recv_bulk(data, urb->transfer_buffer, urb->actual_length) < 0) { - BT_ERR("%s corrupted ACL packet", hdev->name); + bt_dev_err(hdev, "corrupted ACL packet"); hdev->stat.err_rx++; } } else if (urb->status == -ENOENT) { @@ -753,8 +753,8 @@ static void btusb_bulk_complete(struct urb *urb) * -ENODEV: device got disconnected */ if (err != -EPERM && err != -ENODEV) - BT_ERR("%s urb %p failed to resubmit (%d)", - hdev->name, urb, -err); + bt_dev_err(hdev, "urb %p failed to resubmit (%d)", + urb, -err); usb_unanchor_urb(urb); } } @@ -795,8 +795,8 @@ static int btusb_submit_bulk_urb(struct hci_dev *hdev, gfp_t mem_flags) err = usb_submit_urb(urb, mem_flags); if (err < 0) { if (err != -EPERM && err != -ENODEV) - BT_ERR("%s urb %p submission failed (%d)", - hdev->name, urb, -err); + bt_dev_err(hdev, "urb %p submission failed (%d)", + urb, -err); usb_unanchor_urb(urb); } @@ -829,7 +829,7 @@ static void btusb_isoc_complete(struct urb *urb) if (btusb_recv_isoc(data, urb->transfer_buffer + offset, length) < 0) { - BT_ERR("%s corrupted SCO packet", hdev->name); + bt_dev_err(hdev, "corrupted SCO packet"); hdev->stat.err_rx++; } } @@ -849,8 +849,8 @@ static void btusb_isoc_complete(struct urb *urb) * -ENODEV: device got disconnected */ if (err != -EPERM && err != -ENODEV) - BT_ERR("%s urb %p failed to resubmit (%d)", - hdev->name, urb, -err); + bt_dev_err(hdev, "urb %p failed to resubmit (%d)", + urb, -err); usb_unanchor_urb(urb); } } @@ -917,8 +917,8 @@ static int btusb_submit_isoc_urb(struct hci_dev *hdev, gfp_t mem_flags) err = usb_submit_urb(urb, mem_flags); if (err < 0) { if (err != -EPERM && err != -ENODEV) - BT_ERR("%s urb %p submission failed (%d)", - hdev->name, urb, -err); + bt_dev_err(hdev, "urb %p submission failed (%d)", + urb, -err); usb_unanchor_urb(urb); } @@ -962,8 +962,8 @@ static void btusb_diag_complete(struct urb *urb) * -ENODEV: device got disconnected */ if (err != -EPERM && err != -ENODEV) - BT_ERR("%s urb %p failed to resubmit (%d)", - hdev->name, urb, -err); + bt_dev_err(hdev, "urb %p failed to resubmit (%d)", + urb, -err); usb_unanchor_urb(urb); } } @@ -1004,8 +1004,8 @@ static int btusb_submit_diag_urb(struct hci_dev *hdev, gfp_t mem_flags) err = usb_submit_urb(urb, mem_flags); if (err < 0) { if (err != -EPERM && err != -ENODEV) - BT_ERR("%s urb %p submission failed (%d)", - hdev->name, urb, -err); + bt_dev_err(hdev, "urb %p submission failed (%d)", + urb, -err); usb_unanchor_urb(urb); } @@ -1266,8 +1266,8 @@ static int submit_tx_urb(struct hci_dev *hdev, struct urb *urb) err = usb_submit_urb(urb, GFP_KERNEL); if (err < 0) { if (err != -EPERM && err != -ENODEV) - BT_ERR("%s urb %p submission failed (%d)", - hdev->name, urb, -err); + bt_dev_err(hdev, "urb %p submission failed (%d)", + urb, -err); kfree(urb->setup_packet); usb_unanchor_urb(urb); } else { @@ -1362,7 +1362,7 @@ static inline int __set_isoc_interface(struct hci_dev *hdev, int altsetting) err = usb_set_interface(data->udev, 1, altsetting); if (err < 0) { - BT_ERR("%s setting interface failed (%d)", hdev->name, -err); + bt_dev_err(hdev, "setting interface failed (%d)", -err); return err; } @@ -1386,7 +1386,7 @@ static inline int __set_isoc_interface(struct hci_dev *hdev, int altsetting) } if (!data->isoc_tx_ep || !data->isoc_rx_ep) { - BT_ERR("%s invalid SCO descriptors", hdev->name); + bt_dev_err(hdev, "invalid SCO descriptors"); return -ENODEV; } @@ -1481,7 +1481,7 @@ static int btusb_setup_bcm92035(struct hci_dev *hdev) skb = __hci_cmd_sync(hdev, 0xfc3b, 1, &val, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) - BT_ERR("BCM92035 command failed (%ld)", -PTR_ERR(skb)); + bt_dev_err(hdev, "BCM92035 command failed (%ld)", PTR_ERR(skb)); else kfree_skb(skb); @@ -1499,12 +1499,12 @@ static int btusb_setup_csr(struct hci_dev *hdev) HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { int err = PTR_ERR(skb); - BT_ERR("%s: CSR: Local version failed (%d)", hdev->name, err); + bt_dev_err(hdev, "CSR: Local version failed (%d)", err); return err; } if (skb->len != sizeof(struct hci_rp_read_local_version)) { - BT_ERR("%s: CSR: Local version length mismatch", hdev->name); + bt_dev_err(hdev, "CSR: Local version length mismatch"); kfree_skb(skb); return -EIO; } @@ -1566,7 +1566,7 @@ static const struct firmware *btusb_setup_intel_get_fw(struct hci_dev *hdev, } } - BT_INFO("%s: Intel Bluetooth firmware file: %s", hdev->name, fwname); + bt_dev_info(hdev, "Intel Bluetooth firmware file: %s", fwname); return fw; } @@ -1722,18 +1722,18 @@ static int btusb_setup_intel(struct hci_dev *hdev) if (err) return err; - BT_INFO("%s: read Intel version: %02x%02x%02x%02x%02x%02x%02x%02x%02x", - hdev->name, ver.hw_platform, ver.hw_variant, ver.hw_revision, - ver.fw_variant, ver.fw_revision, ver.fw_build_num, - ver.fw_build_ww, ver.fw_build_yy, ver.fw_patch_num); + bt_dev_info(hdev, "read Intel version: %02x%02x%02x%02x%02x%02x%02x%02x%02x", + ver.hw_platform, ver.hw_variant, ver.hw_revision, + ver.fw_variant, ver.fw_revision, ver.fw_build_num, + ver.fw_build_ww, ver.fw_build_yy, ver.fw_patch_num); /* fw_patch_num indicates the version of patch the device currently * have. If there is no patch data in the device, it is always 0x00. * So, if it is other than 0x00, no need to patch the device again. */ if (ver.fw_patch_num) { - BT_INFO("%s: Intel device is already patched. patch num: %02x", - hdev->name, ver.fw_patch_num); + bt_dev_info(hdev, "Intel device is already patched. " + "patch num: %02x", ver.fw_patch_num); goto complete; } @@ -1801,8 +1801,7 @@ static int btusb_setup_intel(struct hci_dev *hdev) if (err) return err; - BT_INFO("%s: Intel Bluetooth firmware patch completed and activated", - hdev->name); + bt_dev_info(hdev, "Intel firmware patch completed and activated"); goto complete; @@ -1812,7 +1811,7 @@ exit_mfg_disable: if (err) return err; - BT_INFO("%s: Intel Bluetooth firmware patch completed", hdev->name); + bt_dev_info(hdev, "Intel firmware patch completed"); goto complete; @@ -1826,8 +1825,7 @@ exit_mfg_deactivate: if (err) return err; - BT_INFO("%s: Intel Bluetooth firmware patch completed and deactivated", - hdev->name); + bt_dev_info(hdev, "Intel firmware patch completed and deactivated"); complete: /* Set the event mask for Intel specific vendor events. This enables @@ -2108,24 +2106,24 @@ static int btusb_setup_intel_new(struct hci_dev *hdev) params = (struct intel_boot_params *)skb->data; - BT_INFO("%s: Device revision is %u", hdev->name, - le16_to_cpu(params->dev_revid)); + bt_dev_info(hdev, "Device revision is %u", + le16_to_cpu(params->dev_revid)); - BT_INFO("%s: Secure boot is %s", hdev->name, - params->secure_boot ? "enabled" : "disabled"); + bt_dev_info(hdev, "Secure boot is %s", + params->secure_boot ? "enabled" : "disabled"); - BT_INFO("%s: OTP lock is %s", hdev->name, - params->otp_lock ? "enabled" : "disabled"); + bt_dev_info(hdev, "OTP lock is %s", + params->otp_lock ? "enabled" : "disabled"); - BT_INFO("%s: API lock is %s", hdev->name, - params->api_lock ? "enabled" : "disabled"); + bt_dev_info(hdev, "API lock is %s", + params->api_lock ? "enabled" : "disabled"); - BT_INFO("%s: Debug lock is %s", hdev->name, - params->debug_lock ? "enabled" : "disabled"); + bt_dev_info(hdev, "Debug lock is %s", + params->debug_lock ? "enabled" : "disabled"); - BT_INFO("%s: Minimum firmware build %u week %u %u", hdev->name, - params->min_fw_build_nn, params->min_fw_build_cw, - 2000 + params->min_fw_build_yy); + bt_dev_info(hdev, "Minimum firmware build %u week %u %u", + params->min_fw_build_nn, params->min_fw_build_cw, + 2000 + params->min_fw_build_yy); /* It is required that every single firmware fragment is acknowledged * with a command complete event. If the boot parameters indicate @@ -2142,7 +2140,7 @@ static int btusb_setup_intel_new(struct hci_dev *hdev) * also be no valid address for the operational firmware. */ if (!bacmp(¶ms->otp_bdaddr, BDADDR_ANY)) { - BT_INFO("%s: No device address configured", hdev->name); + bt_dev_info(hdev, "No device address configured"); set_bit(HCI_QUIRK_INVALID_BDADDR, &hdev->quirks); } @@ -2169,7 +2167,7 @@ static int btusb_setup_intel_new(struct hci_dev *hdev) return err; } - BT_INFO("%s: Found device firmware: %s", hdev->name, fwname); + bt_dev_info(hdev, "Found device firmware: %s", fwname); /* Save the DDC file name for later use to apply once the firmware * downloading is done. @@ -2250,7 +2248,7 @@ static int btusb_setup_intel_new(struct hci_dev *hdev) set_bit(BTUSB_FIRMWARE_LOADED, &data->flags); - BT_INFO("%s: Waiting for firmware download to complete", hdev->name); + bt_dev_info(hdev, "Waiting for firmware download to complete"); /* Before switching the device into operational mode and with that * booting the loaded firmware, wait for the bootloader notification @@ -2287,7 +2285,7 @@ static int btusb_setup_intel_new(struct hci_dev *hdev) delta = ktime_sub(rettime, calltime); duration = (unsigned long long) ktime_to_ns(delta) >> 10; - BT_INFO("%s: Firmware loaded in %llu usecs", hdev->name, duration); + bt_dev_info(hdev, "Firmware loaded in %llu usecs", duration); done: release_firmware(fw); @@ -2313,7 +2311,7 @@ done: * 1 second. However if that happens, then just fail the setup * since something went wrong. */ - BT_INFO("%s: Waiting for device to boot", hdev->name); + bt_dev_info(hdev, "Waiting for device to boot"); err = wait_on_bit_timeout(&data->flags, BTUSB_BOOTING, TASK_INTERRUPTIBLE, @@ -2333,7 +2331,7 @@ done: delta = ktime_sub(rettime, calltime); duration = (unsigned long long) ktime_to_ns(delta) >> 10; - BT_INFO("%s: Device booted in %llu usecs", hdev->name, duration); + bt_dev_info(hdev, "Device booted in %llu usecs", duration); clear_bit(BTUSB_BOOTLOADER, &data->flags); @@ -2436,8 +2434,8 @@ static int btusb_set_bdaddr_marvell(struct hci_dev *hdev, skb = __hci_cmd_sync(hdev, 0xfc22, sizeof(buf), buf, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { ret = PTR_ERR(skb); - BT_ERR("%s: changing Marvell device address failed (%ld)", - hdev->name, ret); + bt_dev_err(hdev, "changing Marvell device address failed (%ld)", + ret); return ret; } kfree_skb(skb); @@ -2461,8 +2459,7 @@ static int btusb_set_bdaddr_ath3012(struct hci_dev *hdev, skb = __hci_cmd_sync(hdev, 0xfc0b, sizeof(buf), buf, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { ret = PTR_ERR(skb); - BT_ERR("%s: Change address command failed (%ld)", - hdev->name, ret); + bt_dev_err(hdev, "Change address command failed (%ld)", ret); return ret; } kfree_skb(skb); @@ -2528,7 +2525,7 @@ static int btusb_qca_send_vendor_req(struct hci_dev *hdev, u8 request, err = usb_control_msg(udev, pipe, request, USB_TYPE_VENDOR | USB_DIR_IN, 0, 0, buf, size, USB_CTRL_SET_TIMEOUT); if (err < 0) { - BT_ERR("%s: Failed to access otp area (%d)", hdev->name, err); + bt_dev_err(hdev, "Failed to access otp area (%d)", err); goto done; } @@ -2568,7 +2565,7 @@ static int btusb_setup_qca_download_fw(struct hci_dev *hdev, err = usb_control_msg(udev, pipe, QCA_DFU_DOWNLOAD, USB_TYPE_VENDOR, 0, 0, buf, size, USB_CTRL_SET_TIMEOUT); if (err < 0) { - BT_ERR("%s: Failed to send headers (%d)", hdev->name, err); + bt_dev_err(hdev, "Failed to send headers (%d)", err); goto done; } @@ -2584,13 +2581,13 @@ static int btusb_setup_qca_download_fw(struct hci_dev *hdev, err = usb_bulk_msg(udev, pipe, buf, size, &len, QCA_DFU_TIMEOUT); if (err < 0) { - BT_ERR("%s: Failed to send body at %zd of %zd (%d)", - hdev->name, sent, firmware->size, err); + bt_dev_err(hdev, "Failed to send body at %zd of %zd (%d)", + sent, firmware->size, err); break; } if (size != len) { - BT_ERR("%s: Failed to get bulk buffer", hdev->name); + bt_dev_err(hdev, "Failed to get bulk buffer"); err = -EILSEQ; break; } @@ -2622,24 +2619,23 @@ static int btusb_setup_qca_load_rampatch(struct hci_dev *hdev, err = request_firmware(&fw, fwname, &hdev->dev); if (err) { - BT_ERR("%s: failed to request rampatch file: %s (%d)", - hdev->name, fwname, err); + bt_dev_err(hdev, "failed to request rampatch file: %s (%d)", + fwname, err); return err; } - BT_INFO("%s: using rampatch file: %s", hdev->name, fwname); + bt_dev_info(hdev, "using rampatch file: %s", fwname); rver = (struct qca_rampatch_version *)(fw->data + info->ver_offset); rver_rom = le16_to_cpu(rver->rom_version); rver_patch = le16_to_cpu(rver->patch_version); - BT_INFO("%s: QCA: patch rome 0x%x build 0x%x, firmware rome 0x%x " - "build 0x%x", hdev->name, rver_rom, rver_patch, ver_rom, - ver_patch); + bt_dev_info(hdev, "QCA: patch rome 0x%x build 0x%x, " + "firmware rome 0x%x build 0x%x", + rver_rom, rver_patch, ver_rom, ver_patch); if (rver_rom != ver_rom || rver_patch <= ver_patch) { - BT_ERR("%s: rampatch file version did not match with firmware", - hdev->name); + bt_dev_err(hdev, "rampatch file version did not match with firmware"); err = -EINVAL; goto done; } @@ -2665,12 +2661,12 @@ static int btusb_setup_qca_load_nvm(struct hci_dev *hdev, err = request_firmware(&fw, fwname, &hdev->dev); if (err) { - BT_ERR("%s: failed to request NVM file: %s (%d)", - hdev->name, fwname, err); + bt_dev_err(hdev, "failed to request NVM file: %s (%d)", + fwname, err); return err; } - BT_INFO("%s: using NVM file: %s", hdev->name, fwname); + bt_dev_info(hdev, "using NVM file: %s", fwname); err = btusb_setup_qca_download_fw(hdev, fw, info->nvm_hdr); @@ -2698,8 +2694,7 @@ static int btusb_setup_qca(struct hci_dev *hdev) info = &qca_devices_table[i]; } if (!info) { - BT_ERR("%s: don't support firmware rome 0x%x", hdev->name, - ver_rom); + bt_dev_err(hdev, "don't support firmware rome 0x%x", ver_rom); return -ENODEV; } @@ -2753,7 +2748,7 @@ static inline int __set_diag_interface(struct hci_dev *hdev) } if (!data->diag_tx_ep || !data->diag_rx_ep) { - BT_ERR("%s invalid diagnostic descriptors", hdev->name); + bt_dev_err(hdev, "invalid diagnostic descriptors"); return -ENODEV; } diff --git a/drivers/bluetooth/hci_ath.c b/drivers/bluetooth/hci_ath.c index 5f17ec339fb8..14ae7ee88acb 100644 --- a/drivers/bluetooth/hci_ath.c +++ b/drivers/bluetooth/hci_ath.c @@ -206,7 +206,7 @@ static int ath_recv(struct hci_uart *hu, const void *data, int count) ath_recv_pkts, ARRAY_SIZE(ath_recv_pkts)); if (IS_ERR(ath->rx_skb)) { int err = PTR_ERR(ath->rx_skb); - BT_ERR("%s: Frame reassembly failed (%d)", hu->hdev->name, err); + bt_dev_err(hu->hdev, "Frame reassembly failed (%d)", err); ath->rx_skb = NULL; return err; } diff --git a/drivers/bluetooth/hci_h4.c b/drivers/bluetooth/hci_h4.c index 3b82a87224a9..fb97a3bf069b 100644 --- a/drivers/bluetooth/hci_h4.c +++ b/drivers/bluetooth/hci_h4.c @@ -132,7 +132,7 @@ static int h4_recv(struct hci_uart *hu, const void *data, int count) h4_recv_pkts, ARRAY_SIZE(h4_recv_pkts)); if (IS_ERR(h4->rx_skb)) { int err = PTR_ERR(h4->rx_skb); - BT_ERR("%s: Frame reassembly failed (%d)", hu->hdev->name, err); + bt_dev_err(hu->hdev, "Frame reassembly failed (%d)", err); h4->rx_skb = NULL; return err; } diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c index 4a949bb60394..bbd7db7384e6 100644 --- a/drivers/bluetooth/hci_qca.c +++ b/drivers/bluetooth/hci_qca.c @@ -800,7 +800,7 @@ static int qca_recv(struct hci_uart *hu, const void *data, int count) qca_recv_pkts, ARRAY_SIZE(qca_recv_pkts)); if (IS_ERR(qca->rx_skb)) { int err = PTR_ERR(qca->rx_skb); - BT_ERR("%s: Frame reassembly failed (%d)", hu->hdev->name, err); + bt_dev_err(hu->hdev, "Frame reassembly failed (%d)", err); qca->rx_skb = NULL; return err; } @@ -863,7 +863,7 @@ static int qca_set_baudrate(struct hci_dev *hdev, uint8_t baudrate) skb = bt_skb_alloc(sizeof(cmd), GFP_ATOMIC); if (!skb) { - BT_ERR("Failed to allocate memory for baudrate packet"); + bt_dev_err(hdev, "Failed to allocate baudrate packet"); return -ENOMEM; } @@ -892,7 +892,7 @@ static int qca_setup(struct hci_uart *hu) unsigned int speed, qca_baudrate = QCA_BAUDRATE_115200; int ret; - BT_INFO("%s: ROME setup", hdev->name); + bt_dev_info(hdev, "ROME setup"); /* Patch downloading has to be done without IBS mode */ clear_bit(STATE_IN_BAND_SLEEP_ENABLED, &qca->flags); @@ -917,11 +917,11 @@ static int qca_setup(struct hci_uart *hu) if (speed) { qca_baudrate = qca_get_baudrate_value(speed); - BT_INFO("%s: Set UART speed to %d", hdev->name, speed); + bt_dev_info(hdev, "Set UART speed to %d", speed); ret = qca_set_baudrate(hdev, qca_baudrate); if (ret) { - BT_ERR("%s: Failed to change the baud rate (%d)", - hdev->name, ret); + bt_dev_err(hdev, "Failed to change the baud rate (%d)", + ret); return ret; } hci_uart_set_baudrate(hu, speed); diff --git a/drivers/bluetooth/hci_serdev.c b/drivers/bluetooth/hci_serdev.c index b725ac4f7ff6..71664b22ec9d 100644 --- a/drivers/bluetooth/hci_serdev.c +++ b/drivers/bluetooth/hci_serdev.c @@ -185,7 +185,7 @@ static int hci_uart_setup(struct hci_dev *hdev) if (hu->proto->set_baudrate && speed) { err = hu->proto->set_baudrate(hu, speed); if (err) - BT_ERR("%s: failed to set baudrate", hdev->name); + bt_dev_err(hdev, "Failed to set baudrate"); else serdev_device_set_baudrate(hu->serdev, speed); } @@ -199,14 +199,13 @@ static int hci_uart_setup(struct hci_dev *hdev) skb = __hci_cmd_sync(hdev, HCI_OP_READ_LOCAL_VERSION, 0, NULL, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { - BT_ERR("%s: Reading local version information failed (%ld)", - hdev->name, PTR_ERR(skb)); + bt_dev_err(hdev, "Reading local version info failed (%ld)", + PTR_ERR(skb)); return 0; } if (skb->len != sizeof(*ver)) { - BT_ERR("%s: Event length mismatch for version information", - hdev->name); + bt_dev_err(hdev, "Event length mismatch for version info"); } kfree_skb(skb); diff --git a/include/net/bluetooth/bluetooth.h b/include/net/bluetooth/bluetooth.h index 020142bb9735..e89cff0c4c23 100644 --- a/include/net/bluetooth/bluetooth.h +++ b/include/net/bluetooth/bluetooth.h @@ -147,6 +147,9 @@ void bt_err_ratelimited(const char *fmt, ...); #define bt_dev_dbg(hdev, fmt, ...) \ BT_DBG("%s: " fmt, (hdev)->name, ##__VA_ARGS__) +#define bt_dev_err_ratelimited(hdev, fmt, ...) \ + BT_ERR_RATELIMITED("%s: " fmt, (hdev)->name, ##__VA_ARGS__) + /* Connection and socket states */ enum { BT_CONNECTED = 1, /* Equal to TCP_ESTABLISHED to make net code happy */ diff --git a/net/bluetooth/a2mp.c b/net/bluetooth/a2mp.c index aad994edd3bb..51c2cf2d8923 100644 --- a/net/bluetooth/a2mp.c +++ b/net/bluetooth/a2mp.c @@ -573,7 +573,7 @@ static int a2mp_discphyslink_req(struct amp_mgr *mgr, struct sk_buff *skb, hcon = hci_conn_hash_lookup_ba(hdev, AMP_LINK, &mgr->l2cap_conn->hcon->dst); if (!hcon) { - BT_ERR("No phys link exist"); + bt_dev_err(hdev, "no phys link exist"); rsp.status = A2MP_STATUS_NO_PHYSICAL_LINK_EXISTS; goto clean; } diff --git a/net/bluetooth/amp.c b/net/bluetooth/amp.c index ebcab5bbadd7..78bec8df8525 100644 --- a/net/bluetooth/amp.c +++ b/net/bluetooth/amp.c @@ -187,7 +187,7 @@ int phylink_gen_key(struct hci_conn *conn, u8 *data, u8 *len, u8 *type) /* Legacy key */ if (conn->key_type < 3) { - BT_ERR("Legacy key type %d", conn->key_type); + bt_dev_err(hdev, "legacy key type %d", conn->key_type); return -EACCES; } @@ -207,7 +207,7 @@ int phylink_gen_key(struct hci_conn *conn, u8 *data, u8 *len, u8 *type) /* Derive Generic AMP Link Key (gamp) */ err = hmac_sha256(keybuf, HCI_AMP_LINK_KEY_SIZE, "gamp", 4, gamp_key); if (err) { - BT_ERR("Could not derive Generic AMP Key: err %d", err); + bt_dev_err(hdev, "could not derive Generic AMP Key: err %d", err); return err; } diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 746adcb62259..a9682534c377 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -729,8 +729,8 @@ static void create_le_conn_complete(struct hci_dev *hdev, u8 status, u16 opcode) goto done; } - BT_ERR("HCI request failed to create LE connection: status 0x%2.2x", - status); + bt_dev_err(hdev, "request failed to create LE connection: " + "status 0x%2.2x", status); if (!conn) goto done; diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 6bc679cd3481..40d260f2bea5 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -267,7 +267,7 @@ static int hci_init1_req(struct hci_request *req, unsigned long opt) amp_init1(req); break; default: - BT_ERR("Unknown device type %d", hdev->dev_type); + bt_dev_err(hdev, "Unknown device type %d", hdev->dev_type); break; } @@ -2150,8 +2150,7 @@ static void hci_error_reset(struct work_struct *work) if (hdev->hw_error) hdev->hw_error(hdev, hdev->hw_error_code); else - BT_ERR("%s hardware error 0x%2.2x", hdev->name, - hdev->hw_error_code); + bt_dev_err(hdev, "hardware error 0x%2.2x", hdev->hw_error_code); if (hci_dev_do_close(hdev)) return; @@ -2524,9 +2523,9 @@ static void hci_cmd_timeout(struct work_struct *work) struct hci_command_hdr *sent = (void *) hdev->sent_cmd->data; u16 opcode = __le16_to_cpu(sent->opcode); - BT_ERR("%s command 0x%4.4x tx timeout", hdev->name, opcode); + bt_dev_err(hdev, "command 0x%4.4x tx timeout", opcode); } else { - BT_ERR("%s command tx timeout", hdev->name); + bt_dev_err(hdev, "command tx timeout"); } atomic_set(&hdev->cmd_cnt, 1); @@ -2858,7 +2857,7 @@ struct hci_conn_params *hci_conn_params_add(struct hci_dev *hdev, params = kzalloc(sizeof(*params), GFP_KERNEL); if (!params) { - BT_ERR("Out of memory"); + bt_dev_err(hdev, "out of memory"); return NULL; } @@ -3393,7 +3392,7 @@ static void hci_send_frame(struct hci_dev *hdev, struct sk_buff *skb) err = hdev->send(hdev, skb); if (err < 0) { - BT_ERR("%s sending frame failed (%d)", hdev->name, err); + bt_dev_err(hdev, "sending frame failed (%d)", err); kfree_skb(skb); } } @@ -3408,7 +3407,7 @@ int hci_send_cmd(struct hci_dev *hdev, __u16 opcode, __u32 plen, skb = hci_prepare_cmd(hdev, opcode, plen, param); if (!skb) { - BT_ERR("%s no memory for command", hdev->name); + bt_dev_err(hdev, "no memory for command"); return -ENOMEM; } @@ -3493,7 +3492,7 @@ static void hci_queue_acl(struct hci_chan *chan, struct sk_buff_head *queue, hci_add_acl_hdr(skb, chan->handle, flags); break; default: - BT_ERR("%s unknown dev_type %d", hdev->name, hdev->dev_type); + bt_dev_err(hdev, "unknown dev_type %d", hdev->dev_type); return; } @@ -3618,7 +3617,7 @@ static struct hci_conn *hci_low_sent(struct hci_dev *hdev, __u8 type, break; default: cnt = 0; - BT_ERR("Unknown link type"); + bt_dev_err(hdev, "unknown link type %d", conn->type); } q = cnt / num; @@ -3635,15 +3634,15 @@ static void hci_link_tx_to(struct hci_dev *hdev, __u8 type) struct hci_conn_hash *h = &hdev->conn_hash; struct hci_conn *c; - BT_ERR("%s link tx timeout", hdev->name); + bt_dev_err(hdev, "link tx timeout"); rcu_read_lock(); /* Kill stalled connections */ list_for_each_entry_rcu(c, &h->list, list) { if (c->type == type && c->sent) { - BT_ERR("%s killing stalled connection %pMR", - hdev->name, &c->dst); + bt_dev_err(hdev, "killing stalled connection %pMR", + &c->dst); hci_disconnect(c, HCI_ERROR_REMOTE_USER_TERM); } } @@ -3724,7 +3723,7 @@ static struct hci_chan *hci_chan_sent(struct hci_dev *hdev, __u8 type, break; default: cnt = 0; - BT_ERR("Unknown link type"); + bt_dev_err(hdev, "unknown link type %d", chan->conn->type); } q = cnt / num; @@ -4066,8 +4065,8 @@ static void hci_acldata_packet(struct hci_dev *hdev, struct sk_buff *skb) l2cap_recv_acldata(conn, skb, flags); return; } else { - BT_ERR("%s ACL packet for unknown connection handle %d", - hdev->name, handle); + bt_dev_err(hdev, "ACL packet for unknown connection handle %d", + handle); } kfree_skb(skb); @@ -4097,8 +4096,8 @@ static void hci_scodata_packet(struct hci_dev *hdev, struct sk_buff *skb) sco_recv_scodata(conn, skb); return; } else { - BT_ERR("%s SCO packet for unknown connection handle %d", - hdev->name, handle); + bt_dev_err(hdev, "SCO packet for unknown connection handle %d", + handle); } kfree_skb(skb); diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 0b4dba08a14e..cd3bbb766c24 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -1188,7 +1188,8 @@ static void hci_cc_le_set_scan_enable(struct hci_dev *hdev, break; default: - BT_ERR("Used reserved LE_Scan_Enable param %d", cp->enable); + bt_dev_err(hdev, "use of reserved LE_Scan_Enable param %d", + cp->enable); break; } @@ -1485,7 +1486,7 @@ static void hci_cs_create_conn(struct hci_dev *hdev, __u8 status) conn = hci_conn_add(hdev, ACL_LINK, &cp->bdaddr, HCI_ROLE_MASTER); if (!conn) - BT_ERR("No memory for new connection"); + bt_dev_err(hdev, "no memory for new connection"); } } @@ -2269,7 +2270,7 @@ static void hci_conn_request_evt(struct hci_dev *hdev, struct sk_buff *skb) conn = hci_conn_add(hdev, ev->link_type, &ev->bdaddr, HCI_ROLE_SLAVE); if (!conn) { - BT_ERR("No memory for new connection"); + bt_dev_err(hdev, "no memory for new connection"); hci_dev_unlock(hdev); return; } @@ -2431,7 +2432,7 @@ static void hci_auth_complete_evt(struct hci_dev *hdev, struct sk_buff *skb) if (!hci_conn_ssp_enabled(conn) && test_bit(HCI_CONN_REAUTH_PEND, &conn->flags)) { - BT_INFO("re-auth of legacy device is not possible."); + bt_dev_info(hdev, "re-auth of legacy device is not possible."); } else { set_bit(HCI_CONN_AUTH, &conn->flags); conn->sec_level = conn->pending_sec_level; @@ -2535,8 +2536,7 @@ static void read_enc_key_size_complete(struct hci_dev *hdev, u8 status, BT_DBG("%s status 0x%02x", hdev->name, status); if (!skb || skb->len < sizeof(*rp)) { - BT_ERR("%s invalid HCI Read Encryption Key Size response", - hdev->name); + bt_dev_err(hdev, "invalid read key size response"); return; } @@ -2554,8 +2554,8 @@ static void read_enc_key_size_complete(struct hci_dev *hdev, u8 status, * supported. */ if (rp->status) { - BT_ERR("%s failed to read key size for handle %u", hdev->name, - handle); + bt_dev_err(hdev, "failed to read key size for handle %u", + handle); conn->enc_key_size = HCI_LINK_KEY_SIZE; } else { conn->enc_key_size = rp->key_size; @@ -2664,7 +2664,7 @@ static void hci_encrypt_change_evt(struct hci_dev *hdev, struct sk_buff *skb) hci_req_add(&req, HCI_OP_READ_ENC_KEY_SIZE, sizeof(cp), &cp); if (hci_req_run_skb(&req, read_enc_key_size_complete)) { - BT_ERR("Sending HCI Read Encryption Key Size failed"); + bt_dev_err(hdev, "sending read key size failed"); conn->enc_key_size = HCI_LINK_KEY_SIZE; goto notify; } @@ -3197,7 +3197,7 @@ static void hci_num_comp_pkts_evt(struct hci_dev *hdev, struct sk_buff *skb) int i; if (hdev->flow_ctl_mode != HCI_FLOW_CTL_MODE_PACKET_BASED) { - BT_ERR("Wrong event for mode %d", hdev->flow_ctl_mode); + bt_dev_err(hdev, "wrong event for mode %d", hdev->flow_ctl_mode); return; } @@ -3249,7 +3249,8 @@ static void hci_num_comp_pkts_evt(struct hci_dev *hdev, struct sk_buff *skb) break; default: - BT_ERR("Unknown type %d conn %p", conn->type, conn); + bt_dev_err(hdev, "unknown type %d conn %p", + conn->type, conn); break; } } @@ -3271,7 +3272,7 @@ static struct hci_conn *__hci_conn_lookup_handle(struct hci_dev *hdev, return chan->conn; break; default: - BT_ERR("%s unknown dev_type %d", hdev->name, hdev->dev_type); + bt_dev_err(hdev, "unknown dev_type %d", hdev->dev_type); break; } @@ -3284,7 +3285,7 @@ static void hci_num_comp_blocks_evt(struct hci_dev *hdev, struct sk_buff *skb) int i; if (hdev->flow_ctl_mode != HCI_FLOW_CTL_MODE_BLOCK_BASED) { - BT_ERR("Wrong event for mode %d", hdev->flow_ctl_mode); + bt_dev_err(hdev, "wrong event for mode %d", hdev->flow_ctl_mode); return; } @@ -3320,7 +3321,8 @@ static void hci_num_comp_blocks_evt(struct hci_dev *hdev, struct sk_buff *skb) break; default: - BT_ERR("Unknown type %d conn %p", conn->type, conn); + bt_dev_err(hdev, "unknown type %d conn %p", + conn->type, conn); break; } } @@ -4479,7 +4481,7 @@ static void hci_le_conn_complete_evt(struct hci_dev *hdev, struct sk_buff *skb) if (!conn) { conn = hci_conn_add(hdev, LE_LINK, &ev->bdaddr, ev->role); if (!conn) { - BT_ERR("No memory for new connection"); + bt_dev_err(hdev, "no memory for new connection"); goto unlock; } @@ -4749,8 +4751,8 @@ static void process_adv_report(struct hci_dev *hdev, u8 type, bdaddr_t *bdaddr, case LE_ADV_SCAN_RSP: break; default: - BT_ERR_RATELIMITED("Unknown advertising packet type: 0x%02x", - type); + bt_dev_err_ratelimited(hdev, "unknown advertising packet " + "type: 0x%02x", type); return; } @@ -4769,8 +4771,7 @@ static void process_adv_report(struct hci_dev *hdev, u8 type, bdaddr_t *bdaddr, /* Adjust for actual length */ if (len != real_len) { - BT_ERR_RATELIMITED("%s advertising data length corrected", - hdev->name); + bt_dev_err_ratelimited(hdev, "advertising data len corrected"); len = real_len; } @@ -5192,7 +5193,7 @@ static bool hci_get_cmd_complete(struct hci_dev *hdev, u16 opcode, return false; if (skb->len < sizeof(*hdr)) { - BT_ERR("Too short HCI event"); + bt_dev_err(hdev, "too short HCI event"); return false; } @@ -5206,12 +5207,13 @@ static bool hci_get_cmd_complete(struct hci_dev *hdev, u16 opcode, } if (hdr->evt != HCI_EV_CMD_COMPLETE) { - BT_DBG("Last event is not cmd complete (0x%2.2x)", hdr->evt); + bt_dev_err(hdev, "last event is not cmd complete (0x%2.2x)", + hdr->evt); return false; } if (skb->len < sizeof(*ev)) { - BT_ERR("Too short cmd_complete event"); + bt_dev_err(hdev, "too short cmd_complete event"); return false; } diff --git a/net/bluetooth/hci_request.c b/net/bluetooth/hci_request.c index 7f28d17dc792..abc0f3224dd1 100644 --- a/net/bluetooth/hci_request.c +++ b/net/bluetooth/hci_request.c @@ -336,8 +336,8 @@ void hci_req_add_ev(struct hci_request *req, u16 opcode, u32 plen, skb = hci_prepare_cmd(hdev, opcode, plen, param); if (!skb) { - BT_ERR("%s no memory for command (opcode 0x%4.4x)", - hdev->name, opcode); + bt_dev_err(hdev, "no memory for command (opcode 0x%4.4x)", + opcode); req->err = -ENOMEM; return; } @@ -1426,7 +1426,7 @@ int hci_update_random_address(struct hci_request *req, bool require_privacy, err = smp_generate_rpa(hdev, hdev->irk, &hdev->rpa); if (err < 0) { - BT_ERR("%s failed to generate new RPA", hdev->name); + bt_dev_err(hdev, "failed to generate new RPA"); return err; } @@ -1788,7 +1788,7 @@ int hci_abort_conn(struct hci_conn *conn, u8 reason) err = hci_req_run(&req, abort_conn_complete); if (err && err != -ENODATA) { - BT_ERR("Failed to run HCI request: err %d", err); + bt_dev_err(conn->hdev, "failed to run HCI request: err %d", err); return err; } @@ -1872,7 +1872,8 @@ static void le_scan_disable_work(struct work_struct *work) hci_req_sync(hdev, le_scan_disable, 0, HCI_CMD_TIMEOUT, &status); if (status) { - BT_ERR("Failed to disable LE scan: status 0x%02x", status); + bt_dev_err(hdev, "failed to disable LE scan: status 0x%02x", + status); return; } @@ -1903,7 +1904,7 @@ static void le_scan_disable_work(struct work_struct *work) hci_req_sync(hdev, bredr_inquiry, DISCOV_INTERLEAVED_INQUIRY_LEN, HCI_CMD_TIMEOUT, &status); if (status) { - BT_ERR("Inquiry failed: status 0x%02x", status); + bt_dev_err(hdev, "inquiry failed: status 0x%02x", status); goto discov_stopped; } @@ -1945,7 +1946,8 @@ static void le_scan_restart_work(struct work_struct *work) hci_req_sync(hdev, le_scan_restart, 0, HCI_CMD_TIMEOUT, &status); if (status) { - BT_ERR("Failed to restart LE scan: status %d", status); + bt_dev_err(hdev, "failed to restart LE scan: status %d", + status); return; } diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index aa300f3a0d51..34aaa2340ac8 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -50,7 +50,7 @@ void hci_conn_add_sysfs(struct hci_conn *conn) dev_set_name(&conn->dev, "%s:%d", hdev->name, conn->handle); if (device_add(&conn->dev) < 0) { - BT_ERR("Failed to register connection device"); + bt_dev_err(hdev, "failed to register connection device"); return; } diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 07a3cc29f426..6e9fc86d8daf 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -2159,8 +2159,8 @@ static int load_link_keys(struct sock *sk, struct hci_dev *hdev, void *data, key_count = __le16_to_cpu(cp->key_count); if (key_count > max_key_count) { - BT_ERR("load_link_keys: too big key_count value %u", - key_count); + bt_dev_err(hdev, "load_link_keys: too big key_count value %u", + key_count); return mgmt_cmd_status(sk, hdev->id, MGMT_OP_LOAD_LINK_KEYS, MGMT_STATUS_INVALID_PARAMS); } @@ -2168,8 +2168,8 @@ static int load_link_keys(struct sock *sk, struct hci_dev *hdev, void *data, expected_len = sizeof(*cp) + key_count * sizeof(struct mgmt_link_key_info); if (expected_len != len) { - BT_ERR("load_link_keys: expected %u bytes, got %u bytes", - expected_len, len); + bt_dev_err(hdev, "load_link_keys: expected %u bytes, got %u bytes", + expected_len, len); return mgmt_cmd_status(sk, hdev->id, MGMT_OP_LOAD_LINK_KEYS, MGMT_STATUS_INVALID_PARAMS); } @@ -2561,7 +2561,7 @@ static int pin_code_reply(struct sock *sk, struct hci_dev *hdev, void *data, memcpy(&ncp.addr, &cp->addr, sizeof(ncp.addr)); - BT_ERR("PIN code is not 16 bytes long"); + bt_dev_err(hdev, "PIN code is not 16 bytes long"); err = send_pin_code_neg_reply(sk, hdev, &ncp); if (err >= 0) @@ -3391,7 +3391,8 @@ static int add_remote_oob_data(struct sock *sk, struct hci_dev *hdev, MGMT_OP_ADD_REMOTE_OOB_DATA, status, &cp->addr, sizeof(cp->addr)); } else { - BT_ERR("add_remote_oob_data: invalid length of %u bytes", len); + bt_dev_err(hdev, "add_remote_oob_data: invalid len of %u bytes", + len); err = mgmt_cmd_status(sk, hdev->id, MGMT_OP_ADD_REMOTE_OOB_DATA, MGMT_STATUS_INVALID_PARAMS); } @@ -3604,8 +3605,8 @@ static int start_service_discovery(struct sock *sk, struct hci_dev *hdev, uuid_count = __le16_to_cpu(cp->uuid_count); if (uuid_count > max_uuid_count) { - BT_ERR("service_discovery: too big uuid_count value %u", - uuid_count); + bt_dev_err(hdev, "service_discovery: too big uuid_count value %u", + uuid_count); err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_START_SERVICE_DISCOVERY, MGMT_STATUS_INVALID_PARAMS, &cp->type, @@ -3615,8 +3616,8 @@ static int start_service_discovery(struct sock *sk, struct hci_dev *hdev, expected_len = sizeof(*cp) + uuid_count * 16; if (expected_len != len) { - BT_ERR("service_discovery: expected %u bytes, got %u bytes", - expected_len, len); + bt_dev_err(hdev, "service_discovery: expected %u bytes, got %u bytes", + expected_len, len); err = mgmt_cmd_complete(sk, hdev->id, MGMT_OP_START_SERVICE_DISCOVERY, MGMT_STATUS_INVALID_PARAMS, &cp->type, @@ -3943,7 +3944,7 @@ static void set_advertising_complete(struct hci_dev *hdev, u8 status, err = hci_req_run(&req, enable_advertising_instance); if (err) - BT_ERR("Failed to re-configure advertising"); + bt_dev_err(hdev, "failed to re-configure advertising"); unlock: hci_dev_unlock(hdev); @@ -4664,15 +4665,16 @@ static int load_irks(struct sock *sk, struct hci_dev *hdev, void *cp_data, irk_count = __le16_to_cpu(cp->irk_count); if (irk_count > max_irk_count) { - BT_ERR("load_irks: too big irk_count value %u", irk_count); + bt_dev_err(hdev, "load_irks: too big irk_count value %u", + irk_count); return mgmt_cmd_status(sk, hdev->id, MGMT_OP_LOAD_IRKS, MGMT_STATUS_INVALID_PARAMS); } expected_len = sizeof(*cp) + irk_count * sizeof(struct mgmt_irk_info); if (expected_len != len) { - BT_ERR("load_irks: expected %u bytes, got %u bytes", - expected_len, len); + bt_dev_err(hdev, "load_irks: expected %u bytes, got %u bytes", + expected_len, len); return mgmt_cmd_status(sk, hdev->id, MGMT_OP_LOAD_IRKS, MGMT_STATUS_INVALID_PARAMS); } @@ -4745,7 +4747,8 @@ static int load_long_term_keys(struct sock *sk, struct hci_dev *hdev, key_count = __le16_to_cpu(cp->key_count); if (key_count > max_key_count) { - BT_ERR("load_ltks: too big key_count value %u", key_count); + bt_dev_err(hdev, "load_ltks: too big key_count value %u", + key_count); return mgmt_cmd_status(sk, hdev->id, MGMT_OP_LOAD_LONG_TERM_KEYS, MGMT_STATUS_INVALID_PARAMS); } @@ -4753,8 +4756,8 @@ static int load_long_term_keys(struct sock *sk, struct hci_dev *hdev, expected_len = sizeof(*cp) + key_count * sizeof(struct mgmt_ltk_info); if (expected_len != len) { - BT_ERR("load_keys: expected %u bytes, got %u bytes", - expected_len, len); + bt_dev_err(hdev, "load_keys: expected %u bytes, got %u bytes", + expected_len, len); return mgmt_cmd_status(sk, hdev->id, MGMT_OP_LOAD_LONG_TERM_KEYS, MGMT_STATUS_INVALID_PARAMS); } @@ -4873,14 +4876,15 @@ static void conn_info_refresh_complete(struct hci_dev *hdev, u8 hci_status, } if (!cp) { - BT_ERR("invalid sent_cmd in conn_info response"); + bt_dev_err(hdev, "invalid sent_cmd in conn_info response"); goto unlock; } handle = __le16_to_cpu(cp->handle); conn = hci_conn_hash_lookup_handle(hdev, handle); if (!conn) { - BT_ERR("unknown handle (%d) in conn_info response", handle); + bt_dev_err(hdev, "unknown handle (%d) in conn_info response", + handle); goto unlock; } @@ -5477,8 +5481,8 @@ static int load_conn_param(struct sock *sk, struct hci_dev *hdev, void *data, param_count = __le16_to_cpu(cp->param_count); if (param_count > max_param_count) { - BT_ERR("load_conn_param: too big param_count value %u", - param_count); + bt_dev_err(hdev, "load_conn_param: too big param_count value %u", + param_count); return mgmt_cmd_status(sk, hdev->id, MGMT_OP_LOAD_CONN_PARAM, MGMT_STATUS_INVALID_PARAMS); } @@ -5486,8 +5490,8 @@ static int load_conn_param(struct sock *sk, struct hci_dev *hdev, void *data, expected_len = sizeof(*cp) + param_count * sizeof(struct mgmt_conn_param); if (expected_len != len) { - BT_ERR("load_conn_param: expected %u bytes, got %u bytes", - expected_len, len); + bt_dev_err(hdev, "load_conn_param: expected %u bytes, got %u bytes", + expected_len, len); return mgmt_cmd_status(sk, hdev->id, MGMT_OP_LOAD_CONN_PARAM, MGMT_STATUS_INVALID_PARAMS); } @@ -5512,7 +5516,7 @@ static int load_conn_param(struct sock *sk, struct hci_dev *hdev, void *data, } else if (param->addr.type == BDADDR_LE_RANDOM) { addr_type = ADDR_LE_DEV_RANDOM; } else { - BT_ERR("Ignoring invalid connection parameters"); + bt_dev_err(hdev, "ignoring invalid connection parameters"); continue; } @@ -5525,14 +5529,14 @@ static int load_conn_param(struct sock *sk, struct hci_dev *hdev, void *data, min, max, latency, timeout); if (hci_check_conn_params(min, max, latency, timeout) < 0) { - BT_ERR("Ignoring invalid connection parameters"); + bt_dev_err(hdev, "ignoring invalid connection parameters"); continue; } hci_param = hci_conn_params_add(hdev, ¶m->addr.bdaddr, addr_type); if (!hci_param) { - BT_ERR("Failed to add connection parameters"); + bt_dev_err(hdev, "failed to add connection parameters"); continue; } diff --git a/net/bluetooth/smp.c b/net/bluetooth/smp.c index d41449b9e9d6..01117ae84f1d 100644 --- a/net/bluetooth/smp.c +++ b/net/bluetooth/smp.c @@ -996,7 +996,8 @@ static u8 smp_random(struct smp_chan *smp) return SMP_UNSPECIFIED; if (crypto_memneq(smp->pcnf, confirm, sizeof(smp->pcnf))) { - BT_ERR("Pairing failed (confirmation values mismatch)"); + bt_dev_err(hcon->hdev, "pairing failed " + "(confirmation values mismatch)"); return SMP_CONFIRM_FAILED; } @@ -1210,7 +1211,7 @@ static void sc_generate_ltk(struct smp_chan *smp) key = hci_find_link_key(hdev, &hcon->dst); if (!key) { - BT_ERR("%s No Link Key found to generate LTK", hdev->name); + bt_dev_err(hdev, "no Link Key found to generate LTK"); return; } @@ -2067,11 +2068,11 @@ static int fixup_sc_false_positive(struct smp_chan *smp) return SMP_UNSPECIFIED; if (hci_dev_test_flag(hdev, HCI_SC_ONLY)) { - BT_ERR("Refusing SMP SC -> legacy fallback in SC-only mode"); + bt_dev_err(hdev, "refusing legacy fallback in SC-only mode"); return SMP_UNSPECIFIED; } - BT_ERR("Trying to fall back to legacy SMP"); + bt_dev_err(hdev, "trying to fall back to legacy SMP"); req = (void *) &smp->preq[1]; rsp = (void *) &smp->prsp[1]; @@ -2082,7 +2083,7 @@ static int fixup_sc_false_positive(struct smp_chan *smp) auth = req->auth_req & AUTH_REQ_MASK(hdev); if (tk_request(conn, 0, auth, rsp->io_capability, req->io_capability)) { - BT_ERR("Failed to fall back to legacy SMP"); + bt_dev_err(hdev, "failed to fall back to legacy SMP"); return SMP_UNSPECIFIED; } @@ -2355,7 +2356,7 @@ int smp_conn_security(struct hci_conn *hcon, __u8 sec_level) chan = conn->smp; if (!chan) { - BT_ERR("SMP security requested but not available"); + bt_dev_err(hcon->hdev, "security requested but not available"); return 1; } @@ -2548,7 +2549,7 @@ static int smp_cmd_ident_addr_info(struct l2cap_conn *conn, */ if (!bacmp(&info->bdaddr, BDADDR_ANY) || !hci_is_identity_address(&info->bdaddr, info->addr_type)) { - BT_ERR("Ignoring IRK with no identity address"); + bt_dev_err(hcon->hdev, "ignoring IRK with no identity address"); goto distribute; } @@ -2953,8 +2954,8 @@ done: return err; drop: - BT_ERR("%s unexpected SMP command 0x%02x from %pMR", hcon->hdev->name, - code, &hcon->dst); + bt_dev_err(hcon->hdev, "unexpected SMP command 0x%02x from %pMR", + code, &hcon->dst); kfree_skb(skb); return 0; } @@ -3021,8 +3022,7 @@ static void bredr_pairing(struct l2cap_chan *chan) smp = smp_chan_create(conn); if (!smp) { - BT_ERR("%s unable to create SMP context for BR/EDR", - hdev->name); + bt_dev_err(hdev, "unable to create SMP context for BR/EDR"); return; } -- cgit v1.2.3 From 384c181e3780ddc45e70483e29d84495b484730d Mon Sep 17 00:00:00 2001 From: Amritha Nambiar Date: Fri, 27 Oct 2017 02:35:34 -0700 Subject: net: sched: Identify hardware traffic classes using classid This patch offloads the classid to hardware and uses the classid reserved in the range :ffe0 - :ffef to identify hardware traffic classes reported via dev->num_tc. tcf_result structure contains the class ID of the class to which the packet belongs and is offloaded to hardware via flower filter. A new helper function is introduced to represent HW traffic classes 0 through 15 using the reserved classid values :ffe0 - :ffef. Signed-off-by: Amritha Nambiar Acked-by: Shannon Nelson Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- include/net/pkt_cls.h | 1 + include/net/sch_generic.h | 7 +++++++ net/sched/cls_flower.c | 2 ++ 3 files changed, 10 insertions(+) (limited to 'include/net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index bf73e1675519..37c5ef766655 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -666,6 +666,7 @@ struct tc_cls_flower_offload { struct fl_flow_key *mask; struct fl_flow_key *key; struct tcf_exts *exts; + u32 classid; }; enum tc_matchall_command { diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 07c179dab478..c23e938f5b19 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -411,6 +411,13 @@ qdisc_class_find(const struct Qdisc_class_hash *hash, u32 id) return NULL; } +static inline int tc_classid_to_hwtc(struct net_device *dev, u32 classid) +{ + u32 hwtc = TC_H_MIN(classid) - TC_H_MIN_PRIORITY; + + return (hwtc < netdev_get_num_tc(dev)) ? hwtc : -EINVAL; +} + int qdisc_class_hash_init(struct Qdisc_class_hash *); void qdisc_class_hash_insert(struct Qdisc_class_hash *, struct Qdisc_class_common *); diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c index 35cb6d684e44..c99fa9e5be46 100644 --- a/net/sched/cls_flower.c +++ b/net/sched/cls_flower.c @@ -241,6 +241,7 @@ static int fl_hw_replace_filter(struct tcf_proto *tp, cls_flower.mask = mask; cls_flower.key = &f->mkey; cls_flower.exts = &f->exts; + cls_flower.classid = f->res.classid; err = tc_setup_cb_call(block, &f->exts, TC_SETUP_CLSFLOWER, &cls_flower, skip_sw); @@ -266,6 +267,7 @@ static void fl_hw_update_stats(struct tcf_proto *tp, struct cls_fl_filter *f) cls_flower.command = TC_CLSFLOWER_STATS; cls_flower.cookie = (unsigned long) f; cls_flower.exts = &f->exts; + cls_flower.classid = f->res.classid; tc_setup_cb_call(block, &f->exts, TC_SETUP_CLSFLOWER, &cls_flower, false); -- cgit v1.2.3 From 6c31e5a91fde2e718e59c8a627c56451f88be54c Mon Sep 17 00:00:00 2001 From: David Ahern Date: Fri, 27 Oct 2017 17:37:13 -0700 Subject: net: Add extack to fib_notifier_info Add extack to fib_notifier_info and plumb through stack to call_fib_rule_notifiers, call_fib_entry_notifiers and call_fib6_entry_notifiers. This allows notifer handlers to return messages to user. Signed-off-by: David Ahern Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller --- include/net/fib_notifier.h | 1 + net/core/fib_rules.c | 9 ++++++--- net/ipv4/fib_trie.c | 13 ++++++++----- net/ipv6/ip6_fib.c | 15 +++++++++------ 4 files changed, 24 insertions(+), 14 deletions(-) (limited to 'include/net') diff --git a/include/net/fib_notifier.h b/include/net/fib_notifier.h index 54cd6b839d2f..c91ec732afd6 100644 --- a/include/net/fib_notifier.h +++ b/include/net/fib_notifier.h @@ -9,6 +9,7 @@ struct fib_notifier_info { struct net *net; int family; + struct netlink_ext_ack *extack; }; enum fib_event_type { diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 9a6d97c1d810..fafd0a41e3f7 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -314,10 +314,12 @@ static int call_fib_rule_notifier(struct notifier_block *nb, struct net *net, static int call_fib_rule_notifiers(struct net *net, enum fib_event_type event_type, struct fib_rule *rule, - struct fib_rules_ops *ops) + struct fib_rules_ops *ops, + struct netlink_ext_ack *extack) { struct fib_rule_notifier_info info = { .info.family = ops->family, + .info.extack = extack, .rule = rule, }; @@ -609,7 +611,7 @@ int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr *nlh, if (rule->tun_id) ip_tunnel_need_metadata(); - call_fib_rule_notifiers(net, FIB_EVENT_RULE_ADD, rule, ops); + call_fib_rule_notifiers(net, FIB_EVENT_RULE_ADD, rule, ops, extack); notify_rule_change(RTM_NEWRULE, rule, ops, nlh, NETLINK_CB(skb).portid); flush_route_cache(ops); rules_ops_put(ops); @@ -749,7 +751,8 @@ int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr *nlh, } } - call_fib_rule_notifiers(net, FIB_EVENT_RULE_DEL, rule, ops); + call_fib_rule_notifiers(net, FIB_EVENT_RULE_DEL, rule, ops, + NULL); notify_rule_change(RTM_DELRULE, rule, ops, nlh, NETLINK_CB(skb).portid); fib_rule_put(rule); diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index aaa1ba09afaa..5ddc4aefff12 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -102,9 +102,11 @@ static int call_fib_entry_notifier(struct notifier_block *nb, struct net *net, static int call_fib_entry_notifiers(struct net *net, enum fib_event_type event_type, u32 dst, - int dst_len, struct fib_alias *fa) + int dst_len, struct fib_alias *fa, + struct netlink_ext_ack *extack) { struct fib_entry_notifier_info info = { + .info.extack = extack, .dst = dst, .dst_len = dst_len, .fi = fa->fa_info, @@ -1214,7 +1216,7 @@ int fib_table_insert(struct net *net, struct fib_table *tb, new_fa->fa_default = -1; call_fib_entry_notifiers(net, FIB_EVENT_ENTRY_REPLACE, - key, plen, new_fa); + key, plen, new_fa, extack); rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen, tb->tb_id, &cfg->fc_nlinfo, nlflags); @@ -1269,7 +1271,7 @@ int fib_table_insert(struct net *net, struct fib_table *tb, tb->tb_num_default++; rt_cache_flush(cfg->fc_nlinfo.nl_net); - call_fib_entry_notifiers(net, event, key, plen, new_fa); + call_fib_entry_notifiers(net, event, key, plen, new_fa, extack); rtmsg_fib(RTM_NEWROUTE, htonl(key), new_fa, plen, new_fa->tb_id, &cfg->fc_nlinfo, nlflags); succeeded: @@ -1569,7 +1571,7 @@ int fib_table_delete(struct net *net, struct fib_table *tb, return -ESRCH; call_fib_entry_notifiers(net, FIB_EVENT_ENTRY_DEL, key, plen, - fa_to_delete); + fa_to_delete, extack); rtmsg_fib(RTM_DELROUTE, htonl(key), fa_to_delete, plen, tb->tb_id, &cfg->fc_nlinfo, 0); @@ -1886,7 +1888,8 @@ int fib_table_flush(struct net *net, struct fib_table *tb) call_fib_entry_notifiers(net, FIB_EVENT_ENTRY_DEL, n->key, - KEYLENGTH - fa->fa_slen, fa); + KEYLENGTH - fa->fa_slen, fa, + NULL); hlist_del_rcu(&fa->fa_list); fib_release_info(fa->fa_info); alias_free_mem_rcu(fa); diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 1ada9672d198..2e2804f5823e 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -353,9 +353,11 @@ static int call_fib6_entry_notifier(struct notifier_block *nb, struct net *net, static int call_fib6_entry_notifiers(struct net *net, enum fib_event_type event_type, - struct rt6_info *rt) + struct rt6_info *rt, + struct netlink_ext_ack *extack) { struct fib6_entry_notifier_info info = { + .info.extack = extack, .rt = rt, }; @@ -868,7 +870,8 @@ static void fib6_purge_rt(struct rt6_info *rt, struct fib6_node *fn, */ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt, - struct nl_info *info, struct mx6_config *mxc) + struct nl_info *info, struct mx6_config *mxc, + struct netlink_ext_ack *extack) { struct rt6_info *leaf = rcu_dereference_protected(fn->leaf, lockdep_is_held(&rt->rt6i_table->tb6_lock)); @@ -1011,7 +1014,7 @@ add: rcu_assign_pointer(rt->rt6i_node, fn); rcu_assign_pointer(*ins, rt); call_fib6_entry_notifiers(info->nl_net, FIB_EVENT_ENTRY_ADD, - rt); + rt, extack); if (!info->skip_notify) inet6_rt_notify(RTM_NEWROUTE, rt, info, nlflags); info->nl_net->ipv6.rt6_stats->fib_rt_entries++; @@ -1040,7 +1043,7 @@ add: rt->dst.rt6_next = iter->dst.rt6_next; rcu_assign_pointer(*ins, rt); call_fib6_entry_notifiers(info->nl_net, FIB_EVENT_ENTRY_REPLACE, - rt); + rt, extack); if (!info->skip_notify) inet6_rt_notify(RTM_NEWROUTE, rt, info, NLM_F_REPLACE); if (!(fn->fn_flags & RTN_RTINFO)) { @@ -1225,7 +1228,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, } #endif - err = fib6_add_rt2node(fn, rt, info, mxc); + err = fib6_add_rt2node(fn, rt, info, mxc, extack); if (!err) { fib6_update_sernum_upto_root(rt, sernum); fib6_start_gc(info->nl_net, rt); @@ -1686,7 +1689,7 @@ static void fib6_del_route(struct fib6_table *table, struct fib6_node *fn, fib6_purge_rt(rt, fn, net); - call_fib6_entry_notifiers(net, FIB_EVENT_ENTRY_DEL, rt); + call_fib6_entry_notifiers(net, FIB_EVENT_ENTRY_DEL, rt, NULL); if (!info->skip_notify) inet6_rt_notify(RTM_DELROUTE, rt, info, 0); rt6_release(rt); -- cgit v1.2.3 From da13c59b9936dfedcf9f2203bd29fbf83ad672bf Mon Sep 17 00:00:00 2001 From: Vishwanath Pai Date: Mon, 30 Oct 2017 19:38:52 -0400 Subject: net: display hw address of source machine during ipv6 DAD failure This patch updates the error messages displayed in kernel log to include hwaddress of the source machine that caused ipv6 duplicate address detection failures. Examples: a) When we receive a NA packet from another machine advertising our address: ICMPv6: NA: 34:ab:cd:56:11:e8 advertised our address 2001:db8:: on eth0! b) When we detect DAD failure during address assignment to an interface: IPv6: eth0: IPv6 duplicate address 2001:db8:: used by 34:ab:cd:56:11:e8 detected! v2: Changed %pI6 to %pI6c in ndisc_recv_na() Chaged the v6 address in the commit message to 2001:db8:: Suggested-by: Igor Lubashev Signed-off-by: Vishwanath Pai Acked-by: David Ahern Signed-off-by: David S. Miller --- include/net/addrconf.h | 2 +- net/ipv6/addrconf.c | 6 +++--- net/ipv6/ndisc.c | 9 +++++---- 3 files changed, 9 insertions(+), 8 deletions(-) (limited to 'include/net') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 15b5ffd7253d..2a616ea53956 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -208,7 +208,7 @@ void ipv6_mc_remap(struct inet6_dev *idev); void ipv6_mc_init_dev(struct inet6_dev *idev); void ipv6_mc_destroy_dev(struct inet6_dev *idev); int ipv6_mc_check_mld(struct sk_buff *skb, struct sk_buff **skb_trimmed); -void addrconf_dad_failure(struct inet6_ifaddr *ifp); +void addrconf_dad_failure(struct sk_buff *skb, struct inet6_ifaddr *ifp); bool ipv6_chk_mcast_addr(struct net_device *dev, const struct in6_addr *group, const struct in6_addr *src_addr); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 5a8a10229a07..cfa374c8b54c 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1987,7 +1987,7 @@ static int addrconf_dad_end(struct inet6_ifaddr *ifp) return err; } -void addrconf_dad_failure(struct inet6_ifaddr *ifp) +void addrconf_dad_failure(struct sk_buff *skb, struct inet6_ifaddr *ifp) { struct inet6_dev *idev = ifp->idev; struct net *net = dev_net(ifp->idev->dev); @@ -1997,8 +1997,8 @@ void addrconf_dad_failure(struct inet6_ifaddr *ifp) return; } - net_info_ratelimited("%s: IPv6 duplicate address %pI6c detected!\n", - ifp->idev->dev->name, &ifp->addr); + net_info_ratelimited("%s: IPv6 duplicate address %pI6c used by %pM detected!\n", + ifp->idev->dev->name, &ifp->addr, eth_hdr(skb)->h_source); spin_lock_bh(&ifp->lock); diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 266a530414d7..f9c3ffe04382 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -46,6 +46,7 @@ #endif #include +#include #include #include #include @@ -822,7 +823,7 @@ have_ifp: * who is doing DAD * so fail our DAD process */ - addrconf_dad_failure(ifp); + addrconf_dad_failure(skb, ifp); return; } else { /* @@ -975,7 +976,7 @@ static void ndisc_recv_na(struct sk_buff *skb) if (ifp) { if (skb->pkt_type != PACKET_LOOPBACK && (ifp->flags & IFA_F_TENTATIVE)) { - addrconf_dad_failure(ifp); + addrconf_dad_failure(skb, ifp); return; } /* What should we make now? The advertisement @@ -989,8 +990,8 @@ static void ndisc_recv_na(struct sk_buff *skb) */ if (skb->pkt_type != PACKET_LOOPBACK) ND_PRINTK(1, warn, - "NA: someone advertises our address %pI6 on %s!\n", - &ifp->addr, ifp->idev->dev->name); + "NA: %pM advertised our address %pI6c on %s!\n", + eth_hdr(skb)->h_source, &ifp->addr, ifp->idev->dev->name); in6_ifa_put(ifp); return; } -- cgit v1.2.3 From 0b5a89caee5c9958c18cd933c7f8891e35b21781 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Wed, 1 Nov 2017 11:47:38 +0100 Subject: net: sched: remove unused tc_should_offload helper tc_should_offload is no longer used, remove it. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/pkt_cls.h | 7 ------- 1 file changed, 7 deletions(-) (limited to 'include/net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index 37c5ef766655..108dcdd96421 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -623,13 +623,6 @@ static inline bool tc_skip_hw(u32 flags) return (flags & TCA_CLS_FLAGS_SKIP_HW) ? true : false; } -static inline bool tc_should_offload(const struct net_device *dev, u32 flags) -{ - if (tc_skip_hw(flags)) - return false; - return tc_can_offload(dev); -} - static inline bool tc_skip_sw(u32 flags) { return (flags & TCA_CLS_FLAGS_SKIP_SW) ? true : false; -- cgit v1.2.3 From 70b5aee46782208c14d93b715e9f62f7fec844f1 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Wed, 1 Nov 2017 11:47:41 +0100 Subject: net: sched: remove ndo_setup_tc check from tc_can_offload Since tc_can_offload is always called from block callback or egdev callback, no need to check if ndo_setup_tc exists. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/pkt_cls.h | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) (limited to 'include/net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index 108dcdd96421..d15c40c7bde7 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -611,11 +611,7 @@ struct tc_cls_u32_offload { static inline bool tc_can_offload(const struct net_device *dev) { - if (!(dev->features & NETIF_F_HW_TC)) - return false; - if (!dev->netdev_ops->ndo_setup_tc) - return false; - return true; + return dev->features & NETIF_F_HW_TC; } static inline bool tc_skip_hw(u32 flags) -- cgit v1.2.3 From 47d3d7ac656a1ffb9d0f0d3c845663ed6fd7e78d Mon Sep 17 00:00:00 2001 From: Tom Herbert Date: Mon, 30 Oct 2017 14:16:00 -0700 Subject: ipv6: Implement limits on Hop-by-Hop and Destination options RFC 8200 (IPv6) defines Hop-by-Hop options and Destination options extension headers. Both of these carry a list of TLVs which is only limited by the maximum length of the extension header (2048 bytes). By the spec a host must process all the TLVs in these options, however these could be used as a fairly obvious denial of service attack. I think this could in fact be a significant DOS vector on the Internet, one mitigating factor might be that many FWs drop all packets with EH (and obviously this is only IPv6) so an Internet wide attack might not be so effective (yet!). By my calculation, the worse case packet with TLVs in a standard 1500 byte MTU packet that would be processed by the stack contains 1282 invidual TLVs (including pad TLVS) or 724 two byte TLVs. I wrote a quick test program that floods a whole bunch of these packets to a host and sure enough there is substantial time spent in ip6_parse_tlv. These packets contain nothing but unknown TLVS (that are ignored), TLV padding, and bogus UDP header with zero payload length. 25.38% [kernel] [k] __fib6_clean_all 21.63% [kernel] [k] ip6_parse_tlv 4.21% [kernel] [k] __local_bh_enable_ip 2.18% [kernel] [k] ip6_pol_route.isra.39 1.98% [kernel] [k] fib6_walk_continue 1.88% [kernel] [k] _raw_write_lock_bh 1.65% [kernel] [k] dst_release This patch adds configurable limits to Destination and Hop-by-Hop options. There are three limits that may be set: - Limit the number of options in a Hop-by-Hop or Destination options extension header. - Limit the byte length of a Hop-by-Hop or Destination options extension header. - Disallow unrecognized options in a Hop-by-Hop or Destination options extension header. The limits are set in corresponding sysctls: ipv6.sysctl.max_dst_opts_cnt ipv6.sysctl.max_hbh_opts_cnt ipv6.sysctl.max_dst_opts_len ipv6.sysctl.max_hbh_opts_len If a max_*_opts_cnt is less than zero then unknown TLVs are disallowed. The number of known TLVs that are allowed is the absolute value of this number. If a limit is exceeded when processing an extension header the packet is dropped. Default values are set to 8 for options counts, and set to INT_MAX for maximum length. Note the choice to limit options to 8 is an arbitrary guess (roughly based on the fact that the stack supports three HBH options and just one destination option). These limits have being proposed in draft-ietf-6man-rfc6434-bis. Tested (by Martin Lau) I tested out 1 thread (i.e. one raw_udp process). I changed the net.ipv6.max_dst_(opts|hbh)_number between 8 to 2048. With sysctls setting to 2048, the softirq% is packed to 100%. With 8, the softirq% is almost unnoticable from mpstat. v2; - Code and documention cleanup. - Change references of RFC2460 to be RFC8200. - Add reference to RFC6434-bis where the limits will be in standard. Signed-off-by: Tom Herbert Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 24 ++++++++++++ include/net/ipv6.h | 40 ++++++++++++++++++++ include/net/netns/ipv6.h | 4 ++ net/ipv6/af_inet6.c | 4 ++ net/ipv6/exthdrs.c | 67 ++++++++++++++++++++++++++++------ net/ipv6/sysctl_net_ipv6.c | 32 ++++++++++++++++ 6 files changed, 159 insertions(+), 12 deletions(-) (limited to 'include/net') diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index 77f4de59dc9c..e6661b205f72 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -1385,6 +1385,30 @@ mld_qrv - INTEGER Default: 2 (as specified by RFC3810 9.1) Minimum: 1 (as specified by RFC6636 4.5) +max_dst_opts_cnt - INTEGER + Maximum number of non-padding TLVs allowed in a Destination + options extension header. If this value is less than zero + then unknown options are disallowed and the number of known + TLVs allowed is the absolute value of this number. + Default: 8 + +max_hbh_opts_cnt - INTEGER + Maximum number of non-padding TLVs allowed in a Hop-by-Hop + options extension header. If this value is less than zero + then unknown options are disallowed and the number of known + TLVs allowed is the absolute value of this number. + Default: 8 + +max dst_opts_len - INTEGER + Maximum length allowed for a Destination options extension + header. + Default: INT_MAX (unlimited) + +max hbh_opts_len - INTEGER + Maximum length allowed for a Hop-by-Hop options extension + header. + Default: INT_MAX (unlimited) + IPv6 Fragmentation: ip6frag_high_thresh - INTEGER diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 3cda3b521c36..fb6d67012de6 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -51,6 +51,46 @@ #define IPV6_DEFAULT_HOPLIMIT 64 #define IPV6_DEFAULT_MCASTHOPS 1 +/* Limits on Hop-by-Hop and Destination options. + * + * Per RFC8200 there is no limit on the maximum number or lengths of options in + * Hop-by-Hop or Destination options other then the packet must fit in an MTU. + * We allow configurable limits in order to mitigate potential denial of + * service attacks. + * + * There are three limits that may be set: + * - Limit the number of options in a Hop-by-Hop or Destination options + * extension header + * - Limit the byte length of a Hop-by-Hop or Destination options extension + * header + * - Disallow unknown options + * + * The limits are expressed in corresponding sysctls: + * + * ipv6.sysctl.max_dst_opts_cnt + * ipv6.sysctl.max_hbh_opts_cnt + * ipv6.sysctl.max_dst_opts_len + * ipv6.sysctl.max_hbh_opts_len + * + * max_*_opts_cnt is the number of TLVs that are allowed for Destination + * options or Hop-by-Hop options. If the number is less than zero then unknown + * TLVs are disallowed and the number of known options that are allowed is the + * absolute value. Setting the value to INT_MAX indicates no limit. + * + * max_*_opts_len is the length limit in bytes of a Destination or + * Hop-by-Hop options extension header. Setting the value to INT_MAX + * indicates no length limit. + * + * If a limit is exceeded when processing an extension header the packet is + * silently discarded. + */ + +/* Default limits for Hop-by-Hop and Destination options */ +#define IP6_DEFAULT_MAX_DST_OPTS_CNT 8 +#define IP6_DEFAULT_MAX_HBH_OPTS_CNT 8 +#define IP6_DEFAULT_MAX_DST_OPTS_LEN INT_MAX /* No limit */ +#define IP6_DEFAULT_MAX_HBH_OPTS_LEN INT_MAX /* No limit */ + /* * Addr type * diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index 2ea1ed341ef8..600ba1c1befc 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -37,6 +37,10 @@ struct netns_sysctl_ipv6 { int idgen_delay; int flowlabel_state_ranges; int flowlabel_reflect; + int max_dst_opts_cnt; + int max_hbh_opts_cnt; + int max_dst_opts_len; + int max_hbh_opts_len; }; struct netns_ipv6 { diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index fe5262fd6aa5..c26f71234b9c 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -810,6 +810,10 @@ static int __net_init inet6_net_init(struct net *net) net->ipv6.sysctl.idgen_retries = 3; net->ipv6.sysctl.idgen_delay = 1 * HZ; net->ipv6.sysctl.flowlabel_state_ranges = 0; + net->ipv6.sysctl.max_dst_opts_cnt = IP6_DEFAULT_MAX_DST_OPTS_CNT; + net->ipv6.sysctl.max_hbh_opts_cnt = IP6_DEFAULT_MAX_HBH_OPTS_CNT; + net->ipv6.sysctl.max_dst_opts_len = IP6_DEFAULT_MAX_DST_OPTS_LEN; + net->ipv6.sysctl.max_hbh_opts_len = IP6_DEFAULT_MAX_HBH_OPTS_LEN; atomic_set(&net->ipv6.fib6_sernum, 1); err = ipv6_init_mibs(net); diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index 9f918a770f87..83bd75713535 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -74,8 +74,20 @@ struct tlvtype_proc { /* An unknown option is detected, decide what to do */ -static bool ip6_tlvopt_unknown(struct sk_buff *skb, int optoff) +static bool ip6_tlvopt_unknown(struct sk_buff *skb, int optoff, + bool disallow_unknowns) { + if (disallow_unknowns) { + /* If unknown TLVs are disallowed by configuration + * then always silently drop packet. Note this also + * means no ICMP parameter problem is sent which + * could be a good property to mitigate a reflection DOS + * attack. + */ + + goto drop; + } + switch ((skb_network_header(skb)[optoff] & 0xC0) >> 6) { case 0: /* ignore */ return true; @@ -95,20 +107,30 @@ static bool ip6_tlvopt_unknown(struct sk_buff *skb, int optoff) return false; } +drop: kfree_skb(skb); return false; } /* Parse tlv encoded option header (hop-by-hop or destination) */ -static bool ip6_parse_tlv(const struct tlvtype_proc *procs, struct sk_buff *skb) +static bool ip6_parse_tlv(const struct tlvtype_proc *procs, + struct sk_buff *skb, + int max_count) { - const struct tlvtype_proc *curr; + int len = (skb_transport_header(skb)[1] + 1) << 3; const unsigned char *nh = skb_network_header(skb); int off = skb_network_header_len(skb); - int len = (skb_transport_header(skb)[1] + 1) << 3; + const struct tlvtype_proc *curr; + bool disallow_unknowns = false; + int tlv_count = 0; int padlen = 0; + if (unlikely(max_count < 0)) { + disallow_unknowns = true; + max_count = -max_count; + } + if (skb_transport_offset(skb) + len > skb_headlen(skb)) goto bad; @@ -149,6 +171,11 @@ static bool ip6_parse_tlv(const struct tlvtype_proc *procs, struct sk_buff *skb) default: /* Other TLV code so scan list */ if (optlen > len) goto bad; + + tlv_count++; + if (tlv_count > max_count) + goto bad; + for (curr = procs; curr->type >= 0; curr++) { if (curr->type == nh[off]) { /* type specific length/alignment @@ -159,10 +186,10 @@ static bool ip6_parse_tlv(const struct tlvtype_proc *procs, struct sk_buff *skb) break; } } - if (curr->type < 0) { - if (ip6_tlvopt_unknown(skb, off) == 0) - return false; - } + if (curr->type < 0 && + !ip6_tlvopt_unknown(skb, off, disallow_unknowns)) + return false; + padlen = 0; break; } @@ -258,23 +285,31 @@ static int ipv6_destopt_rcv(struct sk_buff *skb) __u16 dstbuf; #endif struct dst_entry *dst = skb_dst(skb); + struct net *net = dev_net(skb->dev); + int extlen; if (!pskb_may_pull(skb, skb_transport_offset(skb) + 8) || !pskb_may_pull(skb, (skb_transport_offset(skb) + ((skb_transport_header(skb)[1] + 1) << 3)))) { __IP6_INC_STATS(dev_net(dst->dev), ip6_dst_idev(dst), IPSTATS_MIB_INHDRERRORS); +fail_and_free: kfree_skb(skb); return -1; } + extlen = (skb_transport_header(skb)[1] + 1) << 3; + if (extlen > net->ipv6.sysctl.max_dst_opts_len) + goto fail_and_free; + opt->lastopt = opt->dst1 = skb_network_header_len(skb); #if IS_ENABLED(CONFIG_IPV6_MIP6) dstbuf = opt->dst1; #endif - if (ip6_parse_tlv(tlvprocdestopt_lst, skb)) { - skb->transport_header += (skb_transport_header(skb)[1] + 1) << 3; + if (ip6_parse_tlv(tlvprocdestopt_lst, skb, + init_net.ipv6.sysctl.max_dst_opts_cnt)) { + skb->transport_header += extlen; opt = IP6CB(skb); #if IS_ENABLED(CONFIG_IPV6_MIP6) opt->nhoff = dstbuf; @@ -803,6 +838,8 @@ static const struct tlvtype_proc tlvprochopopt_lst[] = { int ipv6_parse_hopopts(struct sk_buff *skb) { struct inet6_skb_parm *opt = IP6CB(skb); + struct net *net = dev_net(skb->dev); + int extlen; /* * skb_network_header(skb) is equal to skb->data, and @@ -813,13 +850,19 @@ int ipv6_parse_hopopts(struct sk_buff *skb) if (!pskb_may_pull(skb, sizeof(struct ipv6hdr) + 8) || !pskb_may_pull(skb, (sizeof(struct ipv6hdr) + ((skb_transport_header(skb)[1] + 1) << 3)))) { +fail_and_free: kfree_skb(skb); return -1; } + extlen = (skb_transport_header(skb)[1] + 1) << 3; + if (extlen > net->ipv6.sysctl.max_hbh_opts_len) + goto fail_and_free; + opt->flags |= IP6SKB_HOPBYHOP; - if (ip6_parse_tlv(tlvprochopopt_lst, skb)) { - skb->transport_header += (skb_transport_header(skb)[1] + 1) << 3; + if (ip6_parse_tlv(tlvprochopopt_lst, skb, + init_net.ipv6.sysctl.max_hbh_opts_cnt)) { + skb->transport_header += extlen; opt = IP6CB(skb); opt->nhoff = sizeof(struct ipv6hdr); return 1; diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c index 6fbf8ae5e52c..4a2f0fd870bc 100644 --- a/net/ipv6/sysctl_net_ipv6.c +++ b/net/ipv6/sysctl_net_ipv6.c @@ -97,6 +97,34 @@ static struct ctl_table ipv6_table_template[] = { .mode = 0644, .proc_handler = proc_dointvec, }, + { + .procname = "max_dst_opts_number", + .data = &init_net.ipv6.sysctl.max_dst_opts_cnt, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, + { + .procname = "max_hbh_opts_number", + .data = &init_net.ipv6.sysctl.max_hbh_opts_cnt, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, + { + .procname = "max_dst_opts_length", + .data = &init_net.ipv6.sysctl.max_dst_opts_len, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, + { + .procname = "max_hbh_length", + .data = &init_net.ipv6.sysctl.max_hbh_opts_len, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; @@ -157,6 +185,10 @@ static int __net_init ipv6_sysctl_net_init(struct net *net) ipv6_table[7].data = &net->ipv6.sysctl.flowlabel_state_ranges; ipv6_table[8].data = &net->ipv6.sysctl.ip_nonlocal_bind; ipv6_table[9].data = &net->ipv6.sysctl.flowlabel_reflect; + ipv6_table[10].data = &net->ipv6.sysctl.max_dst_opts_cnt; + ipv6_table[11].data = &net->ipv6.sysctl.max_hbh_opts_cnt; + ipv6_table[12].data = &net->ipv6.sysctl.max_dst_opts_len; + ipv6_table[13].data = &net->ipv6.sysctl.max_hbh_opts_len; ipv6_route_table = ipv6_route_sysctl_init(net); if (!ipv6_route_table) -- cgit v1.2.3 From 3ae6ec08292f01c6782d1a80be0b2cc675e0ecfc Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Thu, 2 Nov 2017 17:14:05 +0100 Subject: ipv4: Send a netevent whenever multipath hash policy is changed Devices performing IPv4 forwarding need to update their multipath hash policy whenever it is changed. Inform these devices by generating a netevent. Signed-off-by: Ido Schimmel Reviewed-by: Petr Machata Signed-off-by: Jiri Pirko Acked-by: David Ahern Signed-off-by: David S. Miller --- include/net/netevent.h | 1 + net/ipv4/sysctl_net_ipv4.c | 20 +++++++++++++++++++- 2 files changed, 20 insertions(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/netevent.h b/include/net/netevent.h index f440df172b56..e3f0e8f2f6e8 100644 --- a/include/net/netevent.h +++ b/include/net/netevent.h @@ -25,6 +25,7 @@ enum netevent_notif_type { NETEVENT_NEIGH_UPDATE = 1, /* arg is struct neighbour ptr */ NETEVENT_REDIRECT, /* arg is struct netevent_redirect ptr */ NETEVENT_DELAY_PROBE_TIME_UPDATE, /* arg is struct neigh_parms ptr */ + NETEVENT_MULTIPATH_HASH_UPDATE, /* arg is struct net ptr */ }; int register_netevent_notifier(struct notifier_block *nb); diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 4602af6d5358..8dcc2b185fcc 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -25,6 +25,7 @@ #include #include #include +#include static int zero; static int one = 1; @@ -385,6 +386,23 @@ static int proc_tcp_available_ulp(struct ctl_table *ctl, return ret; } +#ifdef CONFIG_IP_ROUTE_MULTIPATH +static int proc_fib_multipath_hash_policy(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, + loff_t *ppos) +{ + struct net *net = container_of(table->data, struct net, + ipv4.sysctl_fib_multipath_hash_policy); + int ret; + + ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); + if (write && ret == 0) + call_netevent_notifiers(NETEVENT_MULTIPATH_HASH_UPDATE, net); + + return ret; +} +#endif + static struct ctl_table ipv4_table[] = { { .procname = "tcp_max_orphans", @@ -907,7 +925,7 @@ static struct ctl_table ipv4_net_table[] = { .data = &init_net.ipv4.sysctl_fib_multipath_hash_policy, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_fib_multipath_hash_policy, .extra1 = &zero, .extra2 = &one, }, -- cgit v1.2.3 From c7eb7d7230509ec862d4144f7a831f995bc5d028 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Fri, 3 Nov 2017 11:46:24 +0100 Subject: net: sched: introduce chain_head_change callback Add a callback that is to be called whenever head of the chain changes. Also provide a callback for the default case when the caller gets a block using non-extended getter. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/net/pkt_cls.h | 14 ++++++------ include/net/sch_generic.h | 5 ++++- net/sched/cls_api.c | 54 ++++++++++++++++++++++++++++------------------- net/sched/sch_ingress.c | 36 +++++++++++++++++-------------- 4 files changed, 62 insertions(+), 47 deletions(-) (limited to 'include/net') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index d15c40c7bde7..505d4b71975f 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -26,6 +26,8 @@ enum tcf_block_binder_type { struct tcf_block_ext_info { enum tcf_block_binder_type binder_type; + tcf_chain_head_change_t *chain_head_change; + void *chain_head_change_priv; }; struct tcf_block_cb; @@ -37,12 +39,10 @@ struct tcf_chain *tcf_chain_get(struct tcf_block *block, u32 chain_index, void tcf_chain_put(struct tcf_chain *chain); int tcf_block_get(struct tcf_block **p_block, struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q); -int tcf_block_get_ext(struct tcf_block **p_block, - struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q, +int tcf_block_get_ext(struct tcf_block **p_block, struct Qdisc *q, struct tcf_block_ext_info *ei); void tcf_block_put(struct tcf_block *block); -void tcf_block_put_ext(struct tcf_block *block, - struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q, +void tcf_block_put_ext(struct tcf_block *block, struct Qdisc *q, struct tcf_block_ext_info *ei); static inline struct Qdisc *tcf_block_q(struct tcf_block *block) @@ -82,8 +82,7 @@ int tcf_block_get(struct tcf_block **p_block, } static inline -int tcf_block_get_ext(struct tcf_block **p_block, - struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q, +int tcf_block_get_ext(struct tcf_block **p_block, struct Qdisc *q, struct tcf_block_ext_info *ei) { return 0; @@ -94,8 +93,7 @@ static inline void tcf_block_put(struct tcf_block *block) } static inline -void tcf_block_put_ext(struct tcf_block *block, - struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q, +void tcf_block_put_ext(struct tcf_block *block, struct Qdisc *q, struct tcf_block_ext_info *ei) { } diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index c23e938f5b19..f230269e0bfb 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -260,9 +260,12 @@ struct qdisc_skb_cb { unsigned char data[QDISC_CB_PRIV_LEN]; }; +typedef void tcf_chain_head_change_t(struct tcf_proto *tp_head, void *priv); + struct tcf_chain { struct tcf_proto __rcu *filter_chain; - struct tcf_proto __rcu **p_filter_chain; + tcf_chain_head_change_t *chain_head_change; + void *chain_head_change_priv; struct list_head list; struct tcf_block *block; u32 index; /* chain index */ diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 8d1885abee83..206e19f4fc01 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -195,12 +195,19 @@ static struct tcf_chain *tcf_chain_create(struct tcf_block *block, return chain; } +static void tcf_chain_head_change(struct tcf_chain *chain, + struct tcf_proto *tp_head) +{ + if (chain->chain_head_change) + chain->chain_head_change(tp_head, + chain->chain_head_change_priv); +} + static void tcf_chain_flush(struct tcf_chain *chain) { struct tcf_proto *tp; - if (chain->p_filter_chain) - RCU_INIT_POINTER(*chain->p_filter_chain, NULL); + tcf_chain_head_change(chain, NULL); while ((tp = rtnl_dereference(chain->filter_chain)) != NULL) { RCU_INIT_POINTER(chain->filter_chain, tp->next); tcf_chain_put(chain); @@ -242,13 +249,6 @@ void tcf_chain_put(struct tcf_chain *chain) } EXPORT_SYMBOL(tcf_chain_put); -static void -tcf_chain_filter_chain_ptr_set(struct tcf_chain *chain, - struct tcf_proto __rcu **p_filter_chain) -{ - chain->p_filter_chain = p_filter_chain; -} - static void tcf_block_offload_cmd(struct tcf_block *block, struct Qdisc *q, struct tcf_block_ext_info *ei, enum tc_block_command command) @@ -276,8 +276,7 @@ static void tcf_block_offload_unbind(struct tcf_block *block, struct Qdisc *q, tcf_block_offload_cmd(block, q, ei, TC_BLOCK_UNBIND); } -int tcf_block_get_ext(struct tcf_block **p_block, - struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q, +int tcf_block_get_ext(struct tcf_block **p_block, struct Qdisc *q, struct tcf_block_ext_info *ei) { struct tcf_block *block = kzalloc(sizeof(*block), GFP_KERNEL); @@ -295,7 +294,9 @@ int tcf_block_get_ext(struct tcf_block **p_block, err = -ENOMEM; goto err_chain_create; } - tcf_chain_filter_chain_ptr_set(chain, p_filter_chain); + WARN_ON(!ei->chain_head_change); + chain->chain_head_change = ei->chain_head_change; + chain->chain_head_change_priv = ei->chain_head_change_priv; block->net = qdisc_net(q); block->q = q; tcf_block_offload_bind(block, q, ei); @@ -308,12 +309,23 @@ err_chain_create: } EXPORT_SYMBOL(tcf_block_get_ext); +static void tcf_chain_head_change_dflt(struct tcf_proto *tp_head, void *priv) +{ + struct tcf_proto __rcu **p_filter_chain = priv; + + rcu_assign_pointer(*p_filter_chain, tp_head); +} + int tcf_block_get(struct tcf_block **p_block, struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q) { - struct tcf_block_ext_info ei = {0, }; + struct tcf_block_ext_info ei = { + .chain_head_change = tcf_chain_head_change_dflt, + .chain_head_change_priv = p_filter_chain, + }; - return tcf_block_get_ext(p_block, p_filter_chain, q, &ei); + WARN_ON(!p_filter_chain); + return tcf_block_get_ext(p_block, q, &ei); } EXPORT_SYMBOL(tcf_block_get); @@ -334,8 +346,7 @@ static void tcf_block_put_final(struct work_struct *work) * actions should be all removed after flushing. However, filters are now * destroyed in tc filter workqueue with RTNL lock, they can not race here. */ -void tcf_block_put_ext(struct tcf_block *block, - struct tcf_proto __rcu **p_filter_chain, struct Qdisc *q, +void tcf_block_put_ext(struct tcf_block *block, struct Qdisc *q, struct tcf_block_ext_info *ei) { struct tcf_chain *chain, *tmp; @@ -361,7 +372,7 @@ void tcf_block_put(struct tcf_block *block) if (!block) return; - tcf_block_put_ext(block, NULL, block->q, &ei); + tcf_block_put_ext(block, block->q, &ei); } EXPORT_SYMBOL(tcf_block_put); @@ -537,9 +548,8 @@ static void tcf_chain_tp_insert(struct tcf_chain *chain, struct tcf_chain_info *chain_info, struct tcf_proto *tp) { - if (chain->p_filter_chain && - *chain_info->pprev == chain->filter_chain) - rcu_assign_pointer(*chain->p_filter_chain, tp); + if (*chain_info->pprev == chain->filter_chain) + tcf_chain_head_change(chain, tp); RCU_INIT_POINTER(tp->next, tcf_chain_tp_prev(chain_info)); rcu_assign_pointer(*chain_info->pprev, tp); tcf_chain_hold(chain); @@ -551,8 +561,8 @@ static void tcf_chain_tp_remove(struct tcf_chain *chain, { struct tcf_proto *next = rtnl_dereference(chain_info->next); - if (chain->p_filter_chain && tp == chain->filter_chain) - RCU_INIT_POINTER(*chain->p_filter_chain, next); + if (tp == chain->filter_chain) + tcf_chain_head_change(chain, next); RCU_INIT_POINTER(*chain_info->pprev, next); tcf_chain_put(chain); } diff --git a/net/sched/sch_ingress.c b/net/sched/sch_ingress.c index b599db26d34b..811845815b8c 100644 --- a/net/sched/sch_ingress.c +++ b/net/sched/sch_ingress.c @@ -54,6 +54,13 @@ static struct tcf_block *ingress_tcf_block(struct Qdisc *sch, unsigned long cl) return q->block; } +static void clsact_chain_head_change(struct tcf_proto *tp_head, void *priv) +{ + struct tcf_proto __rcu **p_filter_chain = priv; + + rcu_assign_pointer(*p_filter_chain, tp_head); +} + static int ingress_init(struct Qdisc *sch, struct nlattr *opt) { struct ingress_sched_data *q = qdisc_priv(sch); @@ -61,9 +68,10 @@ static int ingress_init(struct Qdisc *sch, struct nlattr *opt) int err; q->block_info.binder_type = TCF_BLOCK_BINDER_TYPE_CLSACT_INGRESS; + q->block_info.chain_head_change = clsact_chain_head_change; + q->block_info.chain_head_change_priv = &dev->ingress_cl_list; - err = tcf_block_get_ext(&q->block, &dev->ingress_cl_list, - sch, &q->block_info); + err = tcf_block_get_ext(&q->block, sch, &q->block_info); if (err) return err; @@ -76,10 +84,8 @@ static int ingress_init(struct Qdisc *sch, struct nlattr *opt) static void ingress_destroy(struct Qdisc *sch) { struct ingress_sched_data *q = qdisc_priv(sch); - struct net_device *dev = qdisc_dev(sch); - tcf_block_put_ext(q->block, &dev->ingress_cl_list, - sch, &q->block_info); + tcf_block_put_ext(q->block, sch, &q->block_info); net_dec_ingress_queue(); } @@ -162,16 +168,18 @@ static int clsact_init(struct Qdisc *sch, struct nlattr *opt) int err; q->ingress_block_info.binder_type = TCF_BLOCK_BINDER_TYPE_CLSACT_INGRESS; + q->ingress_block_info.chain_head_change = clsact_chain_head_change; + q->ingress_block_info.chain_head_change_priv = &dev->ingress_cl_list; - err = tcf_block_get_ext(&q->ingress_block, &dev->ingress_cl_list, - sch, &q->ingress_block_info); + err = tcf_block_get_ext(&q->ingress_block, sch, &q->ingress_block_info); if (err) return err; q->egress_block_info.binder_type = TCF_BLOCK_BINDER_TYPE_CLSACT_EGRESS; + q->egress_block_info.chain_head_change = clsact_chain_head_change; + q->egress_block_info.chain_head_change_priv = &dev->egress_cl_list; - err = tcf_block_get_ext(&q->egress_block, &dev->egress_cl_list, - sch, &q->egress_block_info); + err = tcf_block_get_ext(&q->egress_block, sch, &q->egress_block_info); if (err) goto err_egress_block_get; @@ -183,20 +191,16 @@ static int clsact_init(struct Qdisc *sch, struct nlattr *opt) return 0; err_egress_block_get: - tcf_block_put_ext(q->ingress_block, &dev->ingress_cl_list, - sch, &q->ingress_block_info); + tcf_block_put_ext(q->ingress_block, sch, &q->ingress_block_info); return err; } static void clsact_destroy(struct Qdisc *sch) { struct clsact_sched_data *q = qdisc_priv(sch); - struct net_device *dev = qdisc_dev(sch); - tcf_block_put_ext(q->egress_block, &dev->egress_cl_list, - sch, &q->egress_block_info); - tcf_block_put_ext(q->ingress_block, &dev->ingress_cl_list, - sch, &q->ingress_block_info); + tcf_block_put_ext(q->egress_block, sch, &q->egress_block_info); + tcf_block_put_ext(q->ingress_block, sch, &q->ingress_block_info); net_dec_ingress_queue(); net_dec_egress_queue(); -- cgit v1.2.3 From 46209401f8f6116bd0b2c2d14a63958e83ffca0b Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Fri, 3 Nov 2017 11:46:25 +0100 Subject: net: core: introduce mini_Qdisc and eliminate usage of tp->q for clsact fastpath In sch_handle_egress and sch_handle_ingress tp->q is used only in order to update stats. So stats and filter list are the only things that are needed in clsact qdisc fastpath processing. Introduce new mini_Qdisc struct to hold those items. Also, introduce a helper to swap the mini_Qdisc structures in case filter list head changes. This removes need for tp->q usage without added overhead. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/linux/netdevice.h | 9 ++++++--- include/net/sch_generic.h | 32 ++++++++++++++++++++++++++++++++ net/core/dev.c | 21 +++++++++++---------- net/sched/sch_generic.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++ net/sched/sch_ingress.c | 19 ++++++++++++++----- 5 files changed, 109 insertions(+), 18 deletions(-) (limited to 'include/net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 5e02f79b2110..7de7656550c2 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1559,6 +1559,8 @@ enum netdev_priv_flags { * * @rx_handler: handler for received packets * @rx_handler_data: XXX: need comments on this one + * @miniq_ingress: ingress/clsact qdisc specific data for + * ingress processing * @ingress_queue: XXX: need comments on this one * @broadcast: hw bcast address * @@ -1576,7 +1578,8 @@ enum netdev_priv_flags { * @tx_global_lock: XXX: need comments on this one * * @xps_maps: XXX: need comments on this one - * + * @miniq_egress: clsact qdisc specific data for + * egress processing * @watchdog_timeo: Represents the timeout that is used by * the watchdog (see dev_watchdog()) * @watchdog_timer: List of timers @@ -1795,7 +1798,7 @@ struct net_device { void __rcu *rx_handler_data; #ifdef CONFIG_NET_CLS_ACT - struct tcf_proto __rcu *ingress_cl_list; + struct mini_Qdisc __rcu *miniq_ingress; #endif struct netdev_queue __rcu *ingress_queue; #ifdef CONFIG_NETFILTER_INGRESS @@ -1826,7 +1829,7 @@ struct net_device { struct xps_dev_maps __rcu *xps_maps; #endif #ifdef CONFIG_NET_CLS_ACT - struct tcf_proto __rcu *egress_cl_list; + struct mini_Qdisc __rcu *miniq_egress; #endif /* These may be needed for future network-power-down code. */ diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index f230269e0bfb..c64e62c9450a 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -904,4 +904,36 @@ static inline void psched_ratecfg_getrate(struct tc_ratespec *res, res->linklayer = (r->linklayer & TC_LINKLAYER_MASK); } +/* Mini Qdisc serves for specific needs of ingress/clsact Qdisc. + * The fast path only needs to access filter list and to update stats + */ +struct mini_Qdisc { + struct tcf_proto *filter_list; + struct gnet_stats_basic_cpu __percpu *cpu_bstats; + struct gnet_stats_queue __percpu *cpu_qstats; + struct rcu_head rcu; +}; + +static inline void mini_qdisc_bstats_cpu_update(struct mini_Qdisc *miniq, + const struct sk_buff *skb) +{ + bstats_cpu_update(this_cpu_ptr(miniq->cpu_bstats), skb); +} + +static inline void mini_qdisc_qstats_cpu_drop(struct mini_Qdisc *miniq) +{ + this_cpu_inc(miniq->cpu_qstats->drops); +} + +struct mini_Qdisc_pair { + struct mini_Qdisc miniq1; + struct mini_Qdisc miniq2; + struct mini_Qdisc __rcu **p_miniq; +}; + +void mini_qdisc_pair_swap(struct mini_Qdisc_pair *miniqp, + struct tcf_proto *tp_head); +void mini_qdisc_pair_init(struct mini_Qdisc_pair *miniqp, struct Qdisc *qdisc, + struct mini_Qdisc __rcu **p_miniq); + #endif diff --git a/net/core/dev.c b/net/core/dev.c index 24ac9083bc13..1423cf4d695c 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3274,22 +3274,22 @@ EXPORT_SYMBOL(dev_loopback_xmit); static struct sk_buff * sch_handle_egress(struct sk_buff *skb, int *ret, struct net_device *dev) { - struct tcf_proto *cl = rcu_dereference_bh(dev->egress_cl_list); + struct mini_Qdisc *miniq = rcu_dereference_bh(dev->miniq_egress); struct tcf_result cl_res; - if (!cl) + if (!miniq) return skb; /* qdisc_skb_cb(skb)->pkt_len was already set by the caller. */ - qdisc_bstats_cpu_update(cl->q, skb); + mini_qdisc_bstats_cpu_update(miniq, skb); - switch (tcf_classify(skb, cl, &cl_res, false)) { + switch (tcf_classify(skb, miniq->filter_list, &cl_res, false)) { case TC_ACT_OK: case TC_ACT_RECLASSIFY: skb->tc_index = TC_H_MIN(cl_res.classid); break; case TC_ACT_SHOT: - qdisc_qstats_cpu_drop(cl->q); + mini_qdisc_qstats_cpu_drop(miniq); *ret = NET_XMIT_DROP; kfree_skb(skb); return NULL; @@ -4189,7 +4189,7 @@ sch_handle_ingress(struct sk_buff *skb, struct packet_type **pt_prev, int *ret, struct net_device *orig_dev) { #ifdef CONFIG_NET_CLS_ACT - struct tcf_proto *cl = rcu_dereference_bh(skb->dev->ingress_cl_list); + struct mini_Qdisc *miniq = rcu_dereference_bh(skb->dev->miniq_ingress); struct tcf_result cl_res; /* If there's at least one ingress present somewhere (so @@ -4197,8 +4197,9 @@ sch_handle_ingress(struct sk_buff *skb, struct packet_type **pt_prev, int *ret, * that are not configured with an ingress qdisc will bail * out here. */ - if (!cl) + if (!miniq) return skb; + if (*pt_prev) { *ret = deliver_skb(skb, *pt_prev, orig_dev); *pt_prev = NULL; @@ -4206,15 +4207,15 @@ sch_handle_ingress(struct sk_buff *skb, struct packet_type **pt_prev, int *ret, qdisc_skb_cb(skb)->pkt_len = skb->len; skb->tc_at_ingress = 1; - qdisc_bstats_cpu_update(cl->q, skb); + mini_qdisc_bstats_cpu_update(miniq, skb); - switch (tcf_classify(skb, cl, &cl_res, false)) { + switch (tcf_classify(skb, miniq->filter_list, &cl_res, false)) { case TC_ACT_OK: case TC_ACT_RECLASSIFY: skb->tc_index = TC_H_MIN(cl_res.classid); break; case TC_ACT_SHOT: - qdisc_qstats_cpu_drop(cl->q); + mini_qdisc_qstats_cpu_drop(miniq); kfree_skb(skb); return NULL; case TC_ACT_STOLEN: diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index aa74aa42b5d7..3839cbbdc32b 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -1024,3 +1024,49 @@ void psched_ratecfg_precompute(struct psched_ratecfg *r, } } EXPORT_SYMBOL(psched_ratecfg_precompute); + +static void mini_qdisc_rcu_func(struct rcu_head *head) +{ +} + +void mini_qdisc_pair_swap(struct mini_Qdisc_pair *miniqp, + struct tcf_proto *tp_head) +{ + struct mini_Qdisc *miniq_old = rtnl_dereference(*miniqp->p_miniq); + struct mini_Qdisc *miniq; + + if (!tp_head) { + RCU_INIT_POINTER(*miniqp->p_miniq, NULL); + return; + } + + miniq = !miniq_old || miniq_old == &miniqp->miniq2 ? + &miniqp->miniq1 : &miniqp->miniq2; + + /* We need to make sure that readers won't see the miniq + * we are about to modify. So wait until previous call_rcu_bh callback + * is done. + */ + rcu_barrier_bh(); + miniq->filter_list = tp_head; + rcu_assign_pointer(*miniqp->p_miniq, miniq); + + if (miniq_old) + /* This is counterpart of the rcu barrier above. We need to + * block potential new user of miniq_old until all readers + * are not seeing it. + */ + call_rcu_bh(&miniq_old->rcu, mini_qdisc_rcu_func); +} +EXPORT_SYMBOL(mini_qdisc_pair_swap); + +void mini_qdisc_pair_init(struct mini_Qdisc_pair *miniqp, struct Qdisc *qdisc, + struct mini_Qdisc __rcu **p_miniq) +{ + miniqp->miniq1.cpu_bstats = qdisc->cpu_bstats; + miniqp->miniq1.cpu_qstats = qdisc->cpu_qstats; + miniqp->miniq2.cpu_bstats = qdisc->cpu_bstats; + miniqp->miniq2.cpu_qstats = qdisc->cpu_qstats; + miniqp->p_miniq = p_miniq; +} +EXPORT_SYMBOL(mini_qdisc_pair_init); diff --git a/net/sched/sch_ingress.c b/net/sched/sch_ingress.c index 811845815b8c..5ecc38f35d47 100644 --- a/net/sched/sch_ingress.c +++ b/net/sched/sch_ingress.c @@ -21,6 +21,7 @@ struct ingress_sched_data { struct tcf_block *block; struct tcf_block_ext_info block_info; + struct mini_Qdisc_pair miniqp; }; static struct Qdisc *ingress_leaf(struct Qdisc *sch, unsigned long arg) @@ -56,9 +57,9 @@ static struct tcf_block *ingress_tcf_block(struct Qdisc *sch, unsigned long cl) static void clsact_chain_head_change(struct tcf_proto *tp_head, void *priv) { - struct tcf_proto __rcu **p_filter_chain = priv; + struct mini_Qdisc_pair *miniqp = priv; - rcu_assign_pointer(*p_filter_chain, tp_head); + mini_qdisc_pair_swap(miniqp, tp_head); } static int ingress_init(struct Qdisc *sch, struct nlattr *opt) @@ -67,9 +68,11 @@ static int ingress_init(struct Qdisc *sch, struct nlattr *opt) struct net_device *dev = qdisc_dev(sch); int err; + mini_qdisc_pair_init(&q->miniqp, sch, &dev->miniq_ingress); + q->block_info.binder_type = TCF_BLOCK_BINDER_TYPE_CLSACT_INGRESS; q->block_info.chain_head_change = clsact_chain_head_change; - q->block_info.chain_head_change_priv = &dev->ingress_cl_list; + q->block_info.chain_head_change_priv = &q->miniqp; err = tcf_block_get_ext(&q->block, sch, &q->block_info); if (err) @@ -128,6 +131,8 @@ struct clsact_sched_data { struct tcf_block *egress_block; struct tcf_block_ext_info ingress_block_info; struct tcf_block_ext_info egress_block_info; + struct mini_Qdisc_pair miniqp_ingress; + struct mini_Qdisc_pair miniqp_egress; }; static unsigned long clsact_find(struct Qdisc *sch, u32 classid) @@ -167,17 +172,21 @@ static int clsact_init(struct Qdisc *sch, struct nlattr *opt) struct net_device *dev = qdisc_dev(sch); int err; + mini_qdisc_pair_init(&q->miniqp_ingress, sch, &dev->miniq_ingress); + q->ingress_block_info.binder_type = TCF_BLOCK_BINDER_TYPE_CLSACT_INGRESS; q->ingress_block_info.chain_head_change = clsact_chain_head_change; - q->ingress_block_info.chain_head_change_priv = &dev->ingress_cl_list; + q->ingress_block_info.chain_head_change_priv = &q->miniqp_ingress; err = tcf_block_get_ext(&q->ingress_block, sch, &q->ingress_block_info); if (err) return err; + mini_qdisc_pair_init(&q->miniqp_egress, sch, &dev->miniq_egress); + q->egress_block_info.binder_type = TCF_BLOCK_BINDER_TYPE_CLSACT_EGRESS; q->egress_block_info.chain_head_change = clsact_chain_head_change; - q->egress_block_info.chain_head_change_priv = &dev->egress_cl_list; + q->egress_block_info.chain_head_change_priv = &q->miniqp_egress; err = tcf_block_get_ext(&q->egress_block, sch, &q->egress_block_info); if (err) -- cgit v1.2.3 From 27c565ae9d554fa1c00c799754cff43476c8d3b5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 4 Nov 2017 08:53:27 -0700 Subject: ipv6: remove IN6_ADDR_HSIZE from addrconf.h IN6_ADDR_HSIZE is private to addrconf.c, move it here to avoid confusion. Signed-off-by: Eric Dumazet Acked-by: David Ahern Signed-off-by: David S. Miller --- include/net/addrconf.h | 3 --- net/ipv6/addrconf.c | 2 ++ 2 files changed, 2 insertions(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 3357332ea375..b623b65a79d1 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -59,9 +59,6 @@ struct in6_validator_info { struct netlink_ext_ack *extack; }; -#define IN6_ADDR_HSIZE_SHIFT 8 -#define IN6_ADDR_HSIZE (1 << IN6_ADDR_HSIZE_SHIFT) - int addrconf_init(void); void addrconf_cleanup(void); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 69b8cdb43aa2..66d8c3d912fd 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -157,6 +157,8 @@ static int ipv6_generate_stable_address(struct in6_addr *addr, u8 dad_count, const struct inet6_dev *idev); +#define IN6_ADDR_HSIZE_SHIFT 8 +#define IN6_ADDR_HSIZE (1 << IN6_ADDR_HSIZE_SHIFT) /* * Configured unicast address hash table */ -- cgit v1.2.3 From 99feaafcdb566e8f032e7acc2a303713ad6bf196 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Fri, 3 Nov 2017 19:05:20 -0400 Subject: net: dsa: make switch index unsigned Define the DSA switch index as an unsigned int, because it will never be less than 0. Signed-off-by: Vivien Didelot Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- include/net/dsa.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 50e276dc4c01..fa1c21ab8092 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -209,7 +209,7 @@ struct dsa_switch { * Parent switch tree, and switch index. */ struct dsa_switch_tree *dst; - int index; + unsigned int index; /* Listener for switch fabric events */ struct notifier_block nb; -- cgit v1.2.3 From 49463b7f2da1a115404b02c5533bc2c2125833a3 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Fri, 3 Nov 2017 19:05:21 -0400 Subject: net: dsa: make tree index unsigned Similarly to a DSA switch and port, rename the tree index from "tree" to "index" and make it an unsigned int because it isn't supposed to be less than 0. u32 is an OF specific data used to retrieve the value and has no need to be propagated up to the tree index. Signed-off-by: Vivien Didelot Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- include/net/dsa.h | 2 +- net/dsa/dsa2.c | 14 +++++++------- net/dsa/slave.c | 2 +- 3 files changed, 9 insertions(+), 9 deletions(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index fa1c21ab8092..e54332968417 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -116,7 +116,7 @@ struct dsa_switch_tree { struct raw_notifier_head nh; /* Tree identifier */ - u32 tree; + unsigned int index; /* Number of switches attached to this tree */ struct kref refcount; diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index 797d1156b4e6..8b68dc2f5707 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -27,12 +27,12 @@ static DEFINE_MUTEX(dsa2_mutex); static const struct devlink_ops dsa_devlink_ops = { }; -static struct dsa_switch_tree *dsa_get_dst(u32 tree) +static struct dsa_switch_tree *dsa_get_dst(unsigned int index) { struct dsa_switch_tree *dst; list_for_each_entry(dst, &dsa_switch_trees, list) - if (dst->tree == tree) { + if (dst->index == index) { kref_get(&dst->refcount); return dst; } @@ -53,14 +53,14 @@ static void dsa_put_dst(struct dsa_switch_tree *dst) kref_put(&dst->refcount, dsa_free_dst); } -static struct dsa_switch_tree *dsa_add_dst(u32 tree) +static struct dsa_switch_tree *dsa_add_dst(unsigned int index) { struct dsa_switch_tree *dst; dst = kzalloc(sizeof(*dst), GFP_KERNEL); if (!dst) return NULL; - dst->tree = tree; + dst->index = index; INIT_LIST_HEAD(&dst->list); list_add_tail(&dsa_switch_trees, &dst->list); kref_init(&dst->refcount); @@ -454,7 +454,7 @@ static void dsa_dst_unapply(struct dsa_switch_tree *dst) dst->cpu_dp = NULL; - pr_info("DSA: tree %d unapplied\n", dst->tree); + pr_info("DSA: tree %d unapplied\n", dst->index); dst->applied = false; } @@ -504,7 +504,7 @@ static int dsa_ds_parse(struct dsa_switch_tree *dst, struct dsa_switch *ds) } - pr_info("DSA: switch %d %d parsed\n", dst->tree, ds->index); + pr_info("DSA: switch %d %d parsed\n", dst->index, ds->index); return 0; } @@ -549,7 +549,7 @@ static int dsa_dst_parse(struct dsa_switch_tree *dst) } } - pr_info("DSA: tree %d parsed\n", dst->tree); + pr_info("DSA: tree %d parsed\n", dst->index); return 0; } diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 9b75d0ac4092..814ced75a0cc 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -55,7 +55,7 @@ void dsa_slave_mii_bus_init(struct dsa_switch *ds) ds->slave_mii_bus->read = dsa_slave_phy_read; ds->slave_mii_bus->write = dsa_slave_phy_write; snprintf(ds->slave_mii_bus->id, MII_BUS_ID_SIZE, "dsa-%d.%d", - ds->dst->tree, ds->index); + ds->dst->index, ds->index); ds->slave_mii_bus->parent = ds->dev; ds->slave_mii_bus->phy_mask = ~ds->phys_mii_mask; } -- cgit v1.2.3 From 1f2556916d974cfb62b6af51660186b5f58bd869 Mon Sep 17 00:00:00 2001 From: Priyaranjan Jha Date: Fri, 3 Nov 2017 16:38:48 -0700 Subject: tcp: higher throughput under reordering with adaptive RACK reordering wnd Currently TCP RACK loss detection does not work well if packets are being reordered beyond its static reordering window (min_rtt/4).Under such reordering it may falsely trigger loss recoveries and reduce TCP throughput significantly. This patch improves that by increasing and reducing the reordering window based on DSACK, which is now supported in major TCP implementations. It makes RACK's reo_wnd adaptive based on DSACK and no. of recoveries. - If DSACK is received, increment reo_wnd by min_rtt/4 (upper bounded by srtt), since there is possibility that spurious retransmission was due to reordering delay longer than reo_wnd. - Persist the current reo_wnd value for TCP_RACK_RECOVERY_THRESH (16) no. of successful recoveries (accounts for full DSACK-based loss recovery undo). After that, reset it to default (min_rtt/4). - At max, reo_wnd is incremented only once per rtt. So that the new DSACK on which we are reacting, is due to the spurious retx (approx) after the reo_wnd has been updated last time. - reo_wnd is tracked in terms of steps (of min_rtt/4), rather than absolute value to account for change in rtt. In our internal testing, we observed significant increase in throughput, in scenarios where reordering exceeds min_rtt/4 (previous static value). Signed-off-by: Priyaranjan Jha Signed-off-by: Yuchung Cheng Signed-off-by: Neal Cardwell Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 1 + include/linux/tcp.h | 9 +++++-- include/net/tcp.h | 2 ++ net/ipv4/tcp.c | 1 + net/ipv4/tcp_input.c | 7 +++++ net/ipv4/tcp_minisocks.c | 4 +++ net/ipv4/tcp_recovery.c | 48 ++++++++++++++++++++++++++++++++-- 7 files changed, 68 insertions(+), 4 deletions(-) (limited to 'include/net') diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index e6661b205f72..54410a1d4065 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -454,6 +454,7 @@ tcp_recovery - INTEGER RACK: 0x1 enables the RACK loss detection for fast detection of lost retransmissions and tail drops. + RACK: 0x2 makes RACK's reordering window static (min_rtt/4). Default: 0x1 diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 8c431385b272..22f40c96a15b 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -210,8 +210,13 @@ struct tcp_sock { u64 mstamp; /* (Re)sent time of the skb */ u32 rtt_us; /* Associated RTT */ u32 end_seq; /* Ending TCP sequence of the skb */ - u8 advanced; /* mstamp advanced since last lost marking */ - u8 reord; /* reordering detected */ + u32 last_delivered; /* tp->delivered at last reo_wnd adj */ + u8 reo_wnd_steps; /* Allowed reordering window */ +#define TCP_RACK_RECOVERY_THRESH 16 + u8 reo_wnd_persist:5, /* No. of recovery since last adj */ + dsack_seen:1, /* Whether DSACK seen after last adj */ + advanced:1, /* mstamp advanced since last lost marking */ + reord:1; /* reordering detected */ } rack; u16 advmss; /* Advertised MSS */ u32 chrono_start; /* Start time in jiffies of a TCP chrono */ diff --git a/include/net/tcp.h b/include/net/tcp.h index c2bf2a822b10..babfd4da1515 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -246,6 +246,7 @@ extern int sysctl_tcp_wmem[3]; extern int sysctl_tcp_rmem[3]; #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ +#define TCP_RACK_STATIC_REO_WND 0x2 /* Use static RACK reo wnd */ extern atomic_long_t tcp_memory_allocated; extern struct percpu_counter tcp_sockets_allocated; @@ -1901,6 +1902,7 @@ extern void tcp_rack_mark_lost(struct sock *sk); extern void tcp_rack_advance(struct tcp_sock *tp, u8 sacked, u32 end_seq, u64 xmit_time); extern void tcp_rack_reo_timeout(struct sock *sk); +extern void tcp_rack_update_reo_wnd(struct sock *sk, struct rate_sample *rs); /* At how many usecs into the future should the RTO fire? */ static inline s64 tcp_rto_delta_us(const struct sock *sk) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index a7a0f316eb86..c4cb19ed4628 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -447,6 +447,7 @@ void tcp_init_sock(struct sock *sk) tcp_assign_congestion_control(sk); tp->tsoffset = 0; + tp->rack.reo_wnd_steps = 1; sk->sk_state = TCP_CLOSE; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 8393b405ea98..0ada8bfc2ebd 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -856,6 +856,7 @@ void tcp_disable_fack(struct tcp_sock *tp) static void tcp_dsack_seen(struct tcp_sock *tp) { tp->rx_opt.sack_ok |= TCP_DSACK_SEEN; + tp->rack.dsack_seen = 1; } static void tcp_update_reordering(struct sock *sk, const int metric, @@ -2408,6 +2409,8 @@ static bool tcp_try_undo_recovery(struct sock *sk) mib_idx = LINUX_MIB_TCPFULLUNDO; NET_INC_STATS(sock_net(sk), mib_idx); + } else if (tp->rack.reo_wnd_persist) { + tp->rack.reo_wnd_persist--; } if (tp->snd_una == tp->high_seq && tcp_is_reno(tp)) { /* Hold old state until something *above* high_seq @@ -2427,6 +2430,8 @@ static bool tcp_try_undo_dsack(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); if (tp->undo_marker && !tp->undo_retrans) { + tp->rack.reo_wnd_persist = min(TCP_RACK_RECOVERY_THRESH, + tp->rack.reo_wnd_persist + 1); DBGUNDO(sk, "D-SACK"); tcp_undo_cwnd_reduction(sk, false); NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPDSACKUNDO); @@ -3644,6 +3649,8 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) flag |= tcp_clean_rtx_queue(sk, prior_fackets, prior_snd_una, &acked, &sack_state); + tcp_rack_update_reo_wnd(sk, &rs); + if (tp->tlp_high_seq) tcp_process_tlp_ack(sk, ack, flag); /* If needed, reset TLP/RTO timer; RACK may later override this. */ diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 3c65c1a3f944..4bb86580decd 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -551,6 +551,10 @@ struct sock *tcp_create_openreq_child(const struct sock *sk, newtp->syn_data_acked = 0; newtp->rack.mstamp = 0; newtp->rack.advanced = 0; + newtp->rack.reo_wnd_steps = 1; + newtp->rack.last_delivered = 0; + newtp->rack.reo_wnd_persist = 0; + newtp->rack.dsack_seen = 0; __TCP_INC_STATS(sock_net(sk), TCP_MIB_PASSIVEOPENS); } diff --git a/net/ipv4/tcp_recovery.c b/net/ipv4/tcp_recovery.c index ac3e9c6d3a3d..d3ea89020c69 100644 --- a/net/ipv4/tcp_recovery.c +++ b/net/ipv4/tcp_recovery.c @@ -44,6 +44,7 @@ static bool tcp_rack_sent_after(u64 t1, u64 t2, u32 seq1, u32 seq2) static void tcp_rack_detect_loss(struct sock *sk, u32 *reo_timeout) { struct tcp_sock *tp = tcp_sk(sk); + u32 min_rtt = tcp_min_rtt(tp); struct sk_buff *skb, *n; u32 reo_wnd; @@ -54,8 +55,10 @@ static void tcp_rack_detect_loss(struct sock *sk, u32 *reo_timeout) * to queuing or delayed ACKs. */ reo_wnd = 1000; - if ((tp->rack.reord || !tp->lost_out) && tcp_min_rtt(tp) != ~0U) - reo_wnd = max(tcp_min_rtt(tp) >> 2, reo_wnd); + if ((tp->rack.reord || !tp->lost_out) && min_rtt != ~0U) { + reo_wnd = max((min_rtt >> 2) * tp->rack.reo_wnd_steps, reo_wnd); + reo_wnd = min(reo_wnd, tp->srtt_us >> 3); + } list_for_each_entry_safe(skb, n, &tp->tsorted_sent_queue, tcp_tsorted_anchor) { @@ -160,3 +163,44 @@ void tcp_rack_reo_timeout(struct sock *sk) if (inet_csk(sk)->icsk_pending != ICSK_TIME_RETRANS) tcp_rearm_rto(sk); } + +/* Updates the RACK's reo_wnd based on DSACK and no. of recoveries. + * + * If DSACK is received, increment reo_wnd by min_rtt/4 (upper bounded + * by srtt), since there is possibility that spurious retransmission was + * due to reordering delay longer than reo_wnd. + * + * Persist the current reo_wnd value for TCP_RACK_RECOVERY_THRESH (16) + * no. of successful recoveries (accounts for full DSACK-based loss + * recovery undo). After that, reset it to default (min_rtt/4). + * + * At max, reo_wnd is incremented only once per rtt. So that the new + * DSACK on which we are reacting, is due to the spurious retx (approx) + * after the reo_wnd has been updated last time. + * + * reo_wnd is tracked in terms of steps (of min_rtt/4), rather than + * absolute value to account for change in rtt. + */ +void tcp_rack_update_reo_wnd(struct sock *sk, struct rate_sample *rs) +{ + struct tcp_sock *tp = tcp_sk(sk); + + if (sock_net(sk)->ipv4.sysctl_tcp_recovery & TCP_RACK_STATIC_REO_WND || + !rs->prior_delivered) + return; + + /* Disregard DSACK if a rtt has not passed since we adjusted reo_wnd */ + if (before(rs->prior_delivered, tp->rack.last_delivered)) + tp->rack.dsack_seen = 0; + + /* Adjust the reo_wnd if update is pending */ + if (tp->rack.dsack_seen) { + tp->rack.reo_wnd_steps = min_t(u32, 0xFF, + tp->rack.reo_wnd_steps + 1); + tp->rack.dsack_seen = 0; + tp->rack.last_delivered = tp->delivered; + tp->rack.reo_wnd_persist = TCP_RACK_RECOVERY_THRESH; + } else if (!tp->rack.reo_wnd_persist) { + tp->rack.reo_wnd_steps = 1; + } +} -- cgit v1.2.3 From 5caaed151a68ae36aca2981cc245f5960a0a7603 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 2 Nov 2017 19:41:09 +0100 Subject: netfilter: conntrack: don't cache nlattr_tuple_size result in nla_size We currently call ->nlattr_tuple_size() once at register time and cache result in l4proto->nla_size. nla_size is the only member that is written to, avoiding this would allow to make l4proto trackers const. We can use ->nlattr_tuple_size() at run time, and cache result in the individual trackers instead. This is an intermediate step, next patch removes nlattr_size() callback and computes size at compile time, then removes nla_size. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_conntrack_l4proto.h | 4 ++-- net/ipv4/netfilter/nf_conntrack_proto_icmp.c | 9 +++++++-- net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 9 +++++++-- net/netfilter/nf_conntrack_core.c | 9 +++++++-- net/netfilter/nf_conntrack_netlink.c | 10 +++++++--- net/netfilter/nf_conntrack_proto.c | 2 -- net/netfilter/nf_conntrack_proto_tcp.c | 9 +++++++-- 7 files changed, 37 insertions(+), 15 deletions(-) (limited to 'include/net') diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h index e06518874144..46e786ffcf2f 100644 --- a/include/net/netfilter/nf_conntrack_l4proto.h +++ b/include/net/netfilter/nf_conntrack_l4proto.h @@ -74,7 +74,7 @@ struct nf_conntrack_l4proto { int (*tuple_to_nlattr)(struct sk_buff *skb, const struct nf_conntrack_tuple *t); /* Calculate tuple nlattr size */ - int (*nlattr_tuple_size)(void); + unsigned int (*nlattr_tuple_size)(void); int (*nlattr_to_tuple)(struct nlattr *tb[], struct nf_conntrack_tuple *t); const struct nla_policy *nla_policy; @@ -144,7 +144,7 @@ int nf_ct_port_tuple_to_nlattr(struct sk_buff *skb, const struct nf_conntrack_tuple *tuple); int nf_ct_port_nlattr_to_tuple(struct nlattr *tb[], struct nf_conntrack_tuple *t); -int nf_ct_port_nlattr_tuple_size(void); +unsigned int nf_ct_port_nlattr_tuple_size(void); extern const struct nla_policy nf_ct_port_nla_policy[]; #ifdef CONFIG_SYSCTL diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c index 8969420cecc3..1849fedd9b81 100644 --- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c +++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c @@ -258,9 +258,14 @@ static int icmp_nlattr_to_tuple(struct nlattr *tb[], return 0; } -static int icmp_nlattr_tuple_size(void) +static unsigned int icmp_nlattr_tuple_size(void) { - return nla_policy_len(icmp_nla_policy, CTA_PROTO_MAX + 1); + static unsigned int size __read_mostly; + + if (!size) + size = nla_policy_len(icmp_nla_policy, CTA_PROTO_MAX + 1); + + return size; } #endif diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c index dca921df28e1..3ac0d826afc4 100644 --- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c +++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c @@ -259,9 +259,14 @@ static int icmpv6_nlattr_to_tuple(struct nlattr *tb[], return 0; } -static int icmpv6_nlattr_tuple_size(void) +static unsigned int icmpv6_nlattr_tuple_size(void) { - return nla_policy_len(icmpv6_nla_policy, CTA_PROTO_MAX + 1); + static unsigned int size __read_mostly; + + if (!size) + size = nla_policy_len(icmpv6_nla_policy, CTA_PROTO_MAX + 1); + + return size; } #endif diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 28e675150853..0e516947c16f 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -1563,9 +1563,14 @@ int nf_ct_port_nlattr_to_tuple(struct nlattr *tb[], } EXPORT_SYMBOL_GPL(nf_ct_port_nlattr_to_tuple); -int nf_ct_port_nlattr_tuple_size(void) +unsigned int nf_ct_port_nlattr_tuple_size(void) { - return nla_policy_len(nf_ct_port_nla_policy, CTA_PROTO_MAX + 1); + static unsigned int size __read_mostly; + + if (!size) + size = nla_policy_len(nf_ct_port_nla_policy, CTA_PROTO_MAX + 1); + + return size; } EXPORT_SYMBOL_GPL(nf_ct_port_nlattr_tuple_size); #endif diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index de4053d84364..6e0adfefb9ed 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -533,11 +533,11 @@ nla_put_failure: return -1; } -static inline size_t ctnetlink_proto_size(const struct nf_conn *ct) +static size_t ctnetlink_proto_size(const struct nf_conn *ct) { const struct nf_conntrack_l3proto *l3proto; const struct nf_conntrack_l4proto *l4proto; - size_t len; + size_t len, len4 = 0; l3proto = __nf_ct_l3proto_find(nf_ct_l3num(ct)); len = l3proto->nla_size; @@ -545,8 +545,12 @@ static inline size_t ctnetlink_proto_size(const struct nf_conn *ct) l4proto = __nf_ct_l4proto_find(nf_ct_l3num(ct), nf_ct_protonum(ct)); len += l4proto->nla_size; + if (l4proto->nlattr_tuple_size) { + len4 = l4proto->nlattr_tuple_size(); + len4 *= 3u; /* ORIG, REPLY, MASTER */ + } - return len; + return len + len4; } static inline size_t ctnetlink_acct_size(const struct nf_conn *ct) diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index 83f739e9dc08..3b06ff3f2dee 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -398,8 +398,6 @@ int nf_ct_l4proto_register_one(struct nf_conntrack_l4proto *l4proto) l4proto->nla_size = 0; if (l4proto->nlattr_size) l4proto->nla_size += l4proto->nlattr_size(); - if (l4proto->nlattr_tuple_size) - l4proto->nla_size += 3 * l4proto->nlattr_tuple_size(); rcu_assign_pointer(nf_ct_protos[l4proto->l3proto][l4proto->l4proto], l4proto); diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index 8f283294d70f..b12fc07111d0 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -1277,9 +1277,14 @@ static int tcp_nlattr_size(void) + nla_policy_len(tcp_nla_policy, CTA_PROTOINFO_TCP_MAX + 1); } -static int tcp_nlattr_tuple_size(void) +static unsigned int tcp_nlattr_tuple_size(void) { - return nla_policy_len(nf_ct_port_nla_policy, CTA_PROTO_MAX + 1); + static unsigned int size __read_mostly; + + if (!size) + size = nla_policy_len(nf_ct_port_nla_policy, CTA_PROTO_MAX + 1); + + return size; } #endif -- cgit v1.2.3 From ba0e4d9917b43dfa746cbbcb4477da59aae73bd6 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 9 Oct 2017 19:52:28 +0200 Subject: netfilter: nf_tables: get set elements via netlink This patch adds a new get operation to look up for specific elements in a set via netlink interface. You can also use it to check if an interval already exists. Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 5 ++ net/netfilter/nf_tables_api.c | 184 ++++++++++++++++++++++++++------------ net/netfilter/nft_set_bitmap.c | 18 ++++ net/netfilter/nft_set_hash.c | 39 ++++++++ net/netfilter/nft_set_rbtree.c | 73 +++++++++++++++ 5 files changed, 264 insertions(+), 55 deletions(-) (limited to 'include/net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 0f5b12a4ad09..d011e56cc7a9 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -311,6 +311,7 @@ struct nft_expr; * @flush: deactivate element in the next generation * @remove: remove element from set * @walk: iterate over all set elemeennts + * @get: get set elements * @privsize: function to return size of set private data * @init: initialize private data of new set instance * @destroy: destroy private data of set instance @@ -350,6 +351,10 @@ struct nft_set_ops { void (*walk)(const struct nft_ctx *ctx, struct nft_set *set, struct nft_set_iter *iter); + void * (*get)(const struct net *net, + const struct nft_set *set, + const struct nft_set_elem *elem, + unsigned int flags); unsigned int (*privsize)(const struct nlattr * const nla[], const struct nft_set_desc *desc); diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 3b4a0739ee39..1d66be0d8ef7 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -3586,45 +3586,6 @@ static int nf_tables_dump_set_done(struct netlink_callback *cb) return 0; } -static int nf_tables_getsetelem(struct net *net, struct sock *nlsk, - struct sk_buff *skb, const struct nlmsghdr *nlh, - const struct nlattr * const nla[], - struct netlink_ext_ack *extack) -{ - u8 genmask = nft_genmask_cur(net); - const struct nft_set *set; - struct nft_ctx ctx; - int err; - - err = nft_ctx_init_from_elemattr(&ctx, net, skb, nlh, nla, genmask); - if (err < 0) - return err; - - set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET], - genmask); - if (IS_ERR(set)) - return PTR_ERR(set); - - if (nlh->nlmsg_flags & NLM_F_DUMP) { - struct netlink_dump_control c = { - .dump = nf_tables_dump_set, - .done = nf_tables_dump_set_done, - }; - struct nft_set_dump_ctx *dump_ctx; - - dump_ctx = kmalloc(sizeof(*dump_ctx), GFP_KERNEL); - if (!dump_ctx) - return -ENOMEM; - - dump_ctx->set = set; - dump_ctx->ctx = ctx; - - c.data = dump_ctx; - return netlink_dump_start(nlsk, skb, nlh, &c); - } - return -EOPNOTSUPP; -} - static int nf_tables_fill_setelem_info(struct sk_buff *skb, const struct nft_ctx *ctx, u32 seq, u32 portid, int event, u16 flags, @@ -3670,6 +3631,135 @@ nla_put_failure: return -1; } +static int nft_setelem_parse_flags(const struct nft_set *set, + const struct nlattr *attr, u32 *flags) +{ + if (attr == NULL) + return 0; + + *flags = ntohl(nla_get_be32(attr)); + if (*flags & ~NFT_SET_ELEM_INTERVAL_END) + return -EINVAL; + if (!(set->flags & NFT_SET_INTERVAL) && + *flags & NFT_SET_ELEM_INTERVAL_END) + return -EINVAL; + + return 0; +} + +static int nft_get_set_elem(struct nft_ctx *ctx, struct nft_set *set, + const struct nlattr *attr) +{ + struct nlattr *nla[NFTA_SET_ELEM_MAX + 1]; + const struct nft_set_ext *ext; + struct nft_data_desc desc; + struct nft_set_elem elem; + struct sk_buff *skb; + uint32_t flags = 0; + void *priv; + int err; + + err = nla_parse_nested(nla, NFTA_SET_ELEM_MAX, attr, + nft_set_elem_policy, NULL); + if (err < 0) + return err; + + if (!nla[NFTA_SET_ELEM_KEY]) + return -EINVAL; + + err = nft_setelem_parse_flags(set, nla[NFTA_SET_ELEM_FLAGS], &flags); + if (err < 0) + return err; + + err = nft_data_init(ctx, &elem.key.val, sizeof(elem.key), &desc, + nla[NFTA_SET_ELEM_KEY]); + if (err < 0) + return err; + + err = -EINVAL; + if (desc.type != NFT_DATA_VALUE || desc.len != set->klen) + return err; + + priv = set->ops->get(ctx->net, set, &elem, flags); + if (IS_ERR(priv)) + return PTR_ERR(priv); + + elem.priv = priv; + ext = nft_set_elem_ext(set, &elem); + + err = -ENOMEM; + skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + if (skb == NULL) + goto err1; + + err = nf_tables_fill_setelem_info(skb, ctx, ctx->seq, ctx->portid, + NFT_MSG_NEWSETELEM, 0, set, &elem); + if (err < 0) + goto err2; + + err = nfnetlink_unicast(skb, ctx->net, ctx->portid, MSG_DONTWAIT); + /* This avoids a loop in nfnetlink. */ + if (err < 0) + goto err1; + + return 0; +err2: + kfree_skb(skb); +err1: + /* this avoids a loop in nfnetlink. */ + return err == -EAGAIN ? -ENOBUFS : err; +} + +static int nf_tables_getsetelem(struct net *net, struct sock *nlsk, + struct sk_buff *skb, const struct nlmsghdr *nlh, + const struct nlattr * const nla[], + struct netlink_ext_ack *extack) +{ + u8 genmask = nft_genmask_cur(net); + struct nft_set *set; + struct nlattr *attr; + struct nft_ctx ctx; + int rem, err = 0; + + err = nft_ctx_init_from_elemattr(&ctx, net, skb, nlh, nla, genmask); + if (err < 0) + return err; + + set = nf_tables_set_lookup(ctx.table, nla[NFTA_SET_ELEM_LIST_SET], + genmask); + if (IS_ERR(set)) + return PTR_ERR(set); + + if (nlh->nlmsg_flags & NLM_F_DUMP) { + struct netlink_dump_control c = { + .dump = nf_tables_dump_set, + .done = nf_tables_dump_set_done, + }; + struct nft_set_dump_ctx *dump_ctx; + + dump_ctx = kmalloc(sizeof(*dump_ctx), GFP_KERNEL); + if (!dump_ctx) + return -ENOMEM; + + dump_ctx->set = set; + dump_ctx->ctx = ctx; + + c.data = dump_ctx; + return netlink_dump_start(nlsk, skb, nlh, &c); + } + + if (!nla[NFTA_SET_ELEM_LIST_ELEMENTS]) + return -EINVAL; + + nla_for_each_nested(attr, nla[NFTA_SET_ELEM_LIST_ELEMENTS], rem) { + err = nft_get_set_elem(&ctx, set, attr); + if (err < 0) + break; + } + + return err; +} + static void nf_tables_setelem_notify(const struct nft_ctx *ctx, const struct nft_set *set, const struct nft_set_elem *elem, @@ -3770,22 +3860,6 @@ static void nf_tables_set_elem_destroy(const struct nft_set *set, void *elem) kfree(elem); } -static int nft_setelem_parse_flags(const struct nft_set *set, - const struct nlattr *attr, u32 *flags) -{ - if (attr == NULL) - return 0; - - *flags = ntohl(nla_get_be32(attr)); - if (*flags & ~NFT_SET_ELEM_INTERVAL_END) - return -EINVAL; - if (!(set->flags & NFT_SET_INTERVAL) && - *flags & NFT_SET_ELEM_INTERVAL_END) - return -EINVAL; - - return 0; -} - static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set, const struct nlattr *attr, u32 nlmsg_flags) { diff --git a/net/netfilter/nft_set_bitmap.c b/net/netfilter/nft_set_bitmap.c index 734989c40579..45fb2752fb63 100644 --- a/net/netfilter/nft_set_bitmap.c +++ b/net/netfilter/nft_set_bitmap.c @@ -106,6 +106,23 @@ nft_bitmap_elem_find(const struct nft_set *set, struct nft_bitmap_elem *this, return NULL; } +static void *nft_bitmap_get(const struct net *net, const struct nft_set *set, + const struct nft_set_elem *elem, unsigned int flags) +{ + const struct nft_bitmap *priv = nft_set_priv(set); + u8 genmask = nft_genmask_cur(net); + struct nft_bitmap_elem *be; + + list_for_each_entry_rcu(be, &priv->list, head) { + if (memcmp(nft_set_ext_key(&be->ext), elem->key.val.data, set->klen) || + !nft_set_elem_active(&be->ext, genmask)) + continue; + + return be; + } + return ERR_PTR(-ENOENT); +} + static int nft_bitmap_insert(const struct net *net, const struct nft_set *set, const struct nft_set_elem *elem, struct nft_set_ext **ext) @@ -294,6 +311,7 @@ static struct nft_set_ops nft_bitmap_ops __read_mostly = { .activate = nft_bitmap_activate, .lookup = nft_bitmap_lookup, .walk = nft_bitmap_walk, + .get = nft_bitmap_get, }; static struct nft_set_type nft_bitmap_type __read_mostly = { diff --git a/net/netfilter/nft_set_hash.c b/net/netfilter/nft_set_hash.c index 650677f1e539..c68a7e0fcf1e 100644 --- a/net/netfilter/nft_set_hash.c +++ b/net/netfilter/nft_set_hash.c @@ -95,6 +95,24 @@ static bool nft_rhash_lookup(const struct net *net, const struct nft_set *set, return !!he; } +static void *nft_rhash_get(const struct net *net, const struct nft_set *set, + const struct nft_set_elem *elem, unsigned int flags) +{ + struct nft_rhash *priv = nft_set_priv(set); + struct nft_rhash_elem *he; + struct nft_rhash_cmp_arg arg = { + .genmask = nft_genmask_cur(net), + .set = set, + .key = elem->key.val.data, + }; + + he = rhashtable_lookup_fast(&priv->ht, &arg, nft_rhash_params); + if (he != NULL) + return he; + + return ERR_PTR(-ENOENT); +} + static bool nft_rhash_update(struct nft_set *set, const u32 *key, void *(*new)(struct nft_set *, const struct nft_expr *, @@ -409,6 +427,24 @@ static bool nft_hash_lookup(const struct net *net, const struct nft_set *set, return false; } +static void *nft_hash_get(const struct net *net, const struct nft_set *set, + const struct nft_set_elem *elem, unsigned int flags) +{ + struct nft_hash *priv = nft_set_priv(set); + u8 genmask = nft_genmask_cur(net); + struct nft_hash_elem *he; + u32 hash; + + hash = jhash(elem->key.val.data, set->klen, priv->seed); + hash = reciprocal_scale(hash, priv->buckets); + hlist_for_each_entry_rcu(he, &priv->table[hash], node) { + if (!memcmp(nft_set_ext_key(&he->ext), elem->key.val.data, set->klen) && + nft_set_elem_active(&he->ext, genmask)) + return he; + } + return ERR_PTR(-ENOENT); +} + /* nft_hash_select_ops() makes sure key size can be either 2 or 4 bytes . */ static inline u32 nft_hash_key(const u32 *key, u32 klen) { @@ -600,6 +636,7 @@ static struct nft_set_ops nft_rhash_ops __read_mostly = { .lookup = nft_rhash_lookup, .update = nft_rhash_update, .walk = nft_rhash_walk, + .get = nft_rhash_get, .features = NFT_SET_MAP | NFT_SET_OBJECT | NFT_SET_TIMEOUT, }; @@ -617,6 +654,7 @@ static struct nft_set_ops nft_hash_ops __read_mostly = { .remove = nft_hash_remove, .lookup = nft_hash_lookup, .walk = nft_hash_walk, + .get = nft_hash_get, .features = NFT_SET_MAP | NFT_SET_OBJECT, }; @@ -634,6 +672,7 @@ static struct nft_set_ops nft_hash_fast_ops __read_mostly = { .remove = nft_hash_remove, .lookup = nft_hash_lookup_fast, .walk = nft_hash_walk, + .get = nft_hash_get, .features = NFT_SET_MAP | NFT_SET_OBJECT, }; diff --git a/net/netfilter/nft_set_rbtree.c b/net/netfilter/nft_set_rbtree.c index d83a4ec5900d..e6f08bc5f359 100644 --- a/net/netfilter/nft_set_rbtree.c +++ b/net/netfilter/nft_set_rbtree.c @@ -113,6 +113,78 @@ static bool nft_rbtree_lookup(const struct net *net, const struct nft_set *set, return ret; } +static bool __nft_rbtree_get(const struct net *net, const struct nft_set *set, + const u32 *key, struct nft_rbtree_elem **elem, + unsigned int seq, unsigned int flags, u8 genmask) +{ + struct nft_rbtree_elem *rbe, *interval = NULL; + struct nft_rbtree *priv = nft_set_priv(set); + const struct rb_node *parent; + const void *this; + int d; + + parent = rcu_dereference_raw(priv->root.rb_node); + while (parent != NULL) { + if (read_seqcount_retry(&priv->count, seq)) + return false; + + rbe = rb_entry(parent, struct nft_rbtree_elem, node); + + this = nft_set_ext_key(&rbe->ext); + d = memcmp(this, key, set->klen); + if (d < 0) { + parent = rcu_dereference_raw(parent->rb_left); + interval = rbe; + } else if (d > 0) { + parent = rcu_dereference_raw(parent->rb_right); + } else { + if (!nft_set_elem_active(&rbe->ext, genmask)) + parent = rcu_dereference_raw(parent->rb_left); + + if (!nft_set_ext_exists(&rbe->ext, NFT_SET_EXT_FLAGS) || + (*nft_set_ext_flags(&rbe->ext) & NFT_SET_ELEM_INTERVAL_END) == + (flags & NFT_SET_ELEM_INTERVAL_END)) { + *elem = rbe; + return true; + } + return false; + } + } + + if (set->flags & NFT_SET_INTERVAL && interval != NULL && + nft_set_elem_active(&interval->ext, genmask) && + !nft_rbtree_interval_end(interval)) { + *elem = interval; + return true; + } + + return false; +} + +static void *nft_rbtree_get(const struct net *net, const struct nft_set *set, + const struct nft_set_elem *elem, unsigned int flags) +{ + struct nft_rbtree *priv = nft_set_priv(set); + unsigned int seq = read_seqcount_begin(&priv->count); + struct nft_rbtree_elem *rbe = ERR_PTR(-ENOENT); + const u32 *key = (const u32 *)&elem->key.val; + u8 genmask = nft_genmask_cur(net); + bool ret; + + ret = __nft_rbtree_get(net, set, key, &rbe, seq, flags, genmask); + if (ret || !read_seqcount_retry(&priv->count, seq)) + return rbe; + + read_lock_bh(&priv->lock); + seq = read_seqcount_begin(&priv->count); + ret = __nft_rbtree_get(net, set, key, &rbe, seq, flags, genmask); + if (!ret) + rbe = ERR_PTR(-ENOENT); + read_unlock_bh(&priv->lock); + + return rbe; +} + static int __nft_rbtree_insert(const struct net *net, const struct nft_set *set, struct nft_rbtree_elem *new, struct nft_set_ext **ext) @@ -336,6 +408,7 @@ static struct nft_set_ops nft_rbtree_ops __read_mostly = { .activate = nft_rbtree_activate, .lookup = nft_rbtree_lookup, .walk = nft_rbtree_walk, + .get = nft_rbtree_get, .features = NFT_SET_INTERVAL | NFT_SET_MAP | NFT_SET_OBJECT, }; -- cgit v1.2.3 From 602f3baf22188aad24b9a58be3209ab774b97d74 Mon Sep 17 00:00:00 2001 From: Nogah Frankel Date: Mon, 6 Nov 2017 07:23:41 +0100 Subject: net_sch: red: Add offload ability to RED qdisc Add the ability to offload RED qdisc by using ndo_setup_tc. There are four commands for RED offloading: * TC_RED_SET: handles set and change. * TC_RED_DESTROY: handle qdisc destroy. * TC_RED_STATS: update the qdiscs counters (given as reference) * TC_RED_XSTAT: returns red xstats. Whether RED is being offloaded is being determined every time dump action is being called because parent change of this qdisc could change its offload state but doesn't require any RED function to be called. Signed-off-by: Nogah Frankel Signed-off-by: Jiri Pirko Reviewed-by: Simon Horman Signed-off-by: David S. Miller --- include/linux/netdevice.h | 1 + include/net/pkt_cls.h | 30 ++++++++++++++++ include/uapi/linux/pkt_sched.h | 1 + net/sched/sch_red.c | 79 ++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 111 insertions(+) (limited to 'include/net') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index fda527ccb263..71968a2ca9f3 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -777,6 +777,7 @@ enum tc_setup_type { TC_SETUP_CLSBPF, TC_SETUP_BLOCK, TC_SETUP_CBS, + TC_SETUP_QDISC_RED, }; /* These structures hold the attributes of bpf state that are being passed diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index 98fef3221227..03c208d3c922 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -703,4 +703,34 @@ struct tc_cookie { u8 *data; u32 len; }; + +enum tc_red_command { + TC_RED_REPLACE, + TC_RED_DESTROY, + TC_RED_STATS, + TC_RED_XSTATS, +}; + +struct tc_red_qopt_offload_params { + u32 min; + u32 max; + u32 probability; + bool is_ecn; +}; +struct tc_red_qopt_offload_stats { + struct gnet_stats_basic_packed *bstats; + struct gnet_stats_queue *qstats; +}; + +struct tc_red_qopt_offload { + enum tc_red_command command; + u32 handle; + u32 parent; + union { + struct tc_red_qopt_offload_params set; + struct tc_red_qopt_offload_stats stats; + struct red_stats *xstats; + }; +}; + #endif diff --git a/include/uapi/linux/pkt_sched.h b/include/uapi/linux/pkt_sched.h index 5002562868cc..6a2c5ea7e9c4 100644 --- a/include/uapi/linux/pkt_sched.h +++ b/include/uapi/linux/pkt_sched.h @@ -256,6 +256,7 @@ struct tc_red_qopt { #define TC_RED_ECN 1 #define TC_RED_HARDDROP 2 #define TC_RED_ADAPTATIVE 4 +#define TC_RED_OFFLOADED 8 }; struct tc_red_xstats { diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c index fdfdb56aaae2..007dd8ef8aac 100644 --- a/net/sched/sch_red.c +++ b/net/sched/sch_red.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include @@ -148,11 +149,37 @@ static void red_reset(struct Qdisc *sch) red_restart(&q->vars); } +static int red_offload(struct Qdisc *sch, bool enable) +{ + struct red_sched_data *q = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); + struct tc_red_qopt_offload opt = { + .handle = sch->handle, + .parent = sch->parent, + }; + + if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc) + return -EOPNOTSUPP; + + if (enable) { + opt.command = TC_RED_REPLACE; + opt.set.min = q->parms.qth_min >> q->parms.Wlog; + opt.set.max = q->parms.qth_max >> q->parms.Wlog; + opt.set.probability = q->parms.max_P; + opt.set.is_ecn = red_use_ecn(q); + } else { + opt.command = TC_RED_DESTROY; + } + + return dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_RED, &opt); +} + static void red_destroy(struct Qdisc *sch) { struct red_sched_data *q = qdisc_priv(sch); del_timer_sync(&q->adapt_timer); + red_offload(sch, false); qdisc_destroy(q->qdisc); } @@ -219,6 +246,7 @@ static int red_change(struct Qdisc *sch, struct nlattr *opt) red_start_of_idle_period(&q->vars); sch_tree_unlock(sch); + red_offload(sch, true); return 0; } @@ -244,6 +272,33 @@ static int red_init(struct Qdisc *sch, struct nlattr *opt) return red_change(sch, opt); } +static int red_dump_offload(struct Qdisc *sch, struct tc_red_qopt *opt) +{ + struct net_device *dev = qdisc_dev(sch); + struct tc_red_qopt_offload hw_stats = { + .handle = sch->handle, + .parent = sch->parent, + .command = TC_RED_STATS, + .stats.bstats = &sch->bstats, + .stats.qstats = &sch->qstats, + }; + int err; + + opt->flags &= ~TC_RED_OFFLOADED; + if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc) + return 0; + + err = dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_RED, + &hw_stats); + if (err == -EOPNOTSUPP) + return 0; + + if (!err) + opt->flags |= TC_RED_OFFLOADED; + + return err; +} + static int red_dump(struct Qdisc *sch, struct sk_buff *skb) { struct red_sched_data *q = qdisc_priv(sch); @@ -257,8 +312,13 @@ static int red_dump(struct Qdisc *sch, struct sk_buff *skb) .Plog = q->parms.Plog, .Scell_log = q->parms.Scell_log, }; + int err; sch->qstats.backlog = q->qdisc->qstats.backlog; + err = red_dump_offload(sch, &opt); + if (err) + goto nla_put_failure; + opts = nla_nest_start(skb, TCA_OPTIONS); if (opts == NULL) goto nla_put_failure; @@ -275,6 +335,7 @@ nla_put_failure: static int red_dump_stats(struct Qdisc *sch, struct gnet_dump *d) { struct red_sched_data *q = qdisc_priv(sch); + struct net_device *dev = qdisc_dev(sch); struct tc_red_xstats st = { .early = q->stats.prob_drop + q->stats.forced_drop, .pdrop = q->stats.pdrop, @@ -282,6 +343,24 @@ static int red_dump_stats(struct Qdisc *sch, struct gnet_dump *d) .marked = q->stats.prob_mark + q->stats.forced_mark, }; + if (tc_can_offload(dev) && dev->netdev_ops->ndo_setup_tc) { + struct red_stats hw_stats = {0}; + struct tc_red_qopt_offload hw_stats_request = { + .handle = sch->handle, + .parent = sch->parent, + .command = TC_RED_XSTATS, + .xstats = &hw_stats, + }; + if (!dev->netdev_ops->ndo_setup_tc(dev, + TC_SETUP_QDISC_RED, + &hw_stats_request)) { + st.early += hw_stats.prob_drop + hw_stats.forced_drop; + st.pdrop += hw_stats.pdrop; + st.other += hw_stats.other; + st.marked += hw_stats.prob_mark + hw_stats.forced_mark; + } + } + return gnet_stats_copy_app(d, &st, sizeof(st)); } -- cgit v1.2.3 From b2d0f5d5dc53532e6f07bc546a476a55ebdfe0f3 Mon Sep 17 00:00:00 2001 From: Yi Yang Date: Tue, 7 Nov 2017 21:07:02 +0800 Subject: openvswitch: enable NSH support v16->17 - Fixed disputed check code: keep them in nsh_push and nsh_pop but also add them in __ovs_nla_copy_actions v15->v16 - Add csum recalculation for nsh_push, nsh_pop and set_nsh pointed out by Pravin - Move nsh key into the union with ipv4 and ipv6 and add check for nsh key in match_validate pointed out by Pravin - Add nsh check in validate_set and __ovs_nla_copy_actions v14->v15 - Check size in nsh_hdr_from_nlattr - Fixed four small issues pointed out By Jiri and Eric v13->v14 - Rename skb_push_nsh to nsh_push per Dave's comment - Rename skb_pop_nsh to nsh_pop per Dave's comment v12->v13 - Fix NSH header length check in set_nsh v11->v12 - Fix missing changes old comments pointed out - Fix new comments for v11 v10->v11 - Fix the left three disputable comments for v9 but not fixed in v10. v9->v10 - Change struct ovs_key_nsh to struct ovs_nsh_key_base base; __be32 context[NSH_MD1_CONTEXT_SIZE]; - Fix new comments for v9 v8->v9 - Fix build error reported by daily intel build because nsh module isn't selected by openvswitch v7->v8 - Rework nested value and mask for OVS_KEY_ATTR_NSH - Change pop_nsh to adapt to nsh kernel module - Fix many issues per comments from Jiri Benc v6->v7 - Remove NSH GSO patches in v6 because Jiri Benc reworked it as another patch series and they have been merged. - Change it to adapt to nsh kernel module added by NSH GSO patch series v5->v6 - Fix the rest comments for v4. - Add NSH GSO support for VxLAN-gpe + NSH and Eth + NSH. v4->v5 - Fix many comments by Jiri Benc and Eric Garver for v4. v3->v4 - Add new NSH match field ttl - Update NSH header to the latest format which will be final format and won't change per its author's confirmation. - Fix comments for v3. v2->v3 - Change OVS_KEY_ATTR_NSH to nested key to handle length-fixed attributes and length-variable attriubte more flexibly. - Remove struct ovs_action_push_nsh completely - Add code to handle nested attribute for SET_MASKED - Change PUSH_NSH to use the nested OVS_KEY_ATTR_NSH to transfer NSH header data. - Fix comments and coding style issues by Jiri and Eric v1->v2 - Change encap_nsh and decap_nsh to push_nsh and pop_nsh - Dynamically allocate struct ovs_action_push_nsh for length-variable metadata. OVS master and 2.8 branch has merged NSH userspace patch series, this patch is to enable NSH support in kernel data path in order that OVS can support NSH in compat mode by porting this. Signed-off-by: Yi Yang Acked-by: Jiri Benc Acked-by: Eric Garver Acked-by: Pravin Shelar Signed-off-by: David S. Miller --- include/net/nsh.h | 3 + include/uapi/linux/openvswitch.h | 29 ++++ net/nsh/nsh.c | 60 +++++++ net/openvswitch/Kconfig | 1 + net/openvswitch/actions.c | 116 +++++++++++++ net/openvswitch/flow.c | 51 ++++++ net/openvswitch/flow.h | 7 + net/openvswitch/flow_netlink.c | 343 ++++++++++++++++++++++++++++++++++++++- net/openvswitch/flow_netlink.h | 5 + 9 files changed, 613 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/nsh.h b/include/net/nsh.h index a1eaea20be96..350b1ad11c7f 100644 --- a/include/net/nsh.h +++ b/include/net/nsh.h @@ -304,4 +304,7 @@ static inline void nsh_set_flags_ttl_len(struct nshhdr *nsh, u8 flags, NSH_FLAGS_MASK | NSH_TTL_MASK | NSH_LEN_MASK); } +int nsh_push(struct sk_buff *skb, const struct nshhdr *pushed_nh); +int nsh_pop(struct sk_buff *skb); + #endif /* __NET_NSH_H */ diff --git a/include/uapi/linux/openvswitch.h b/include/uapi/linux/openvswitch.h index 501e4c4e2a03..ec75a685f1dd 100644 --- a/include/uapi/linux/openvswitch.h +++ b/include/uapi/linux/openvswitch.h @@ -336,6 +336,7 @@ enum ovs_key_attr { OVS_KEY_ATTR_CT_LABELS, /* 16-octet connection tracking label */ OVS_KEY_ATTR_CT_ORIG_TUPLE_IPV4, /* struct ovs_key_ct_tuple_ipv4 */ OVS_KEY_ATTR_CT_ORIG_TUPLE_IPV6, /* struct ovs_key_ct_tuple_ipv6 */ + OVS_KEY_ATTR_NSH, /* Nested set of ovs_nsh_key_* */ #ifdef __KERNEL__ OVS_KEY_ATTR_TUNNEL_INFO, /* struct ip_tunnel_info */ @@ -495,6 +496,30 @@ struct ovs_key_ct_tuple_ipv6 { __u8 ipv6_proto; }; +enum ovs_nsh_key_attr { + OVS_NSH_KEY_ATTR_UNSPEC, + OVS_NSH_KEY_ATTR_BASE, /* struct ovs_nsh_key_base. */ + OVS_NSH_KEY_ATTR_MD1, /* struct ovs_nsh_key_md1. */ + OVS_NSH_KEY_ATTR_MD2, /* variable-length octets for MD type 2. */ + __OVS_NSH_KEY_ATTR_MAX +}; + +#define OVS_NSH_KEY_ATTR_MAX (__OVS_NSH_KEY_ATTR_MAX - 1) + +struct ovs_nsh_key_base { + __u8 flags; + __u8 ttl; + __u8 mdtype; + __u8 np; + __be32 path_hdr; +}; + +#define NSH_MD1_CONTEXT_SIZE 4 + +struct ovs_nsh_key_md1 { + __be32 context[NSH_MD1_CONTEXT_SIZE]; +}; + /** * enum ovs_flow_attr - attributes for %OVS_FLOW_* commands. * @OVS_FLOW_ATTR_KEY: Nested %OVS_KEY_ATTR_* attributes specifying the flow @@ -811,6 +836,8 @@ struct ovs_action_push_eth { * @OVS_ACTION_ATTR_POP_ETH: Pop the outermost Ethernet header off the * packet. * @OVS_ACTION_ATTR_CT_CLEAR: Clear conntrack state from the packet. + * @OVS_ACTION_ATTR_PUSH_NSH: push NSH header to the packet. + * @OVS_ACTION_ATTR_POP_NSH: pop the outermost NSH header off the packet. * * Only a single header can be set with a single %OVS_ACTION_ATTR_SET. Not all * fields within a header are modifiable, e.g. the IPv4 protocol and fragment @@ -841,6 +868,8 @@ enum ovs_action_attr { OVS_ACTION_ATTR_PUSH_ETH, /* struct ovs_action_push_eth. */ OVS_ACTION_ATTR_POP_ETH, /* No argument. */ OVS_ACTION_ATTR_CT_CLEAR, /* No argument. */ + OVS_ACTION_ATTR_PUSH_NSH, /* Nested OVS_NSH_KEY_ATTR_*. */ + OVS_ACTION_ATTR_POP_NSH, /* No argument. */ __OVS_ACTION_ATTR_MAX, /* Nothing past this will be accepted * from userspace. */ diff --git a/net/nsh/nsh.c b/net/nsh/nsh.c index 58fb827439a8..d7da99a0b0b8 100644 --- a/net/nsh/nsh.c +++ b/net/nsh/nsh.c @@ -14,6 +14,66 @@ #include #include +int nsh_push(struct sk_buff *skb, const struct nshhdr *pushed_nh) +{ + struct nshhdr *nh; + size_t length = nsh_hdr_len(pushed_nh); + u8 next_proto; + + if (skb->mac_len) { + next_proto = TUN_P_ETHERNET; + } else { + next_proto = tun_p_from_eth_p(skb->protocol); + if (!next_proto) + return -EAFNOSUPPORT; + } + + /* Add the NSH header */ + if (skb_cow_head(skb, length) < 0) + return -ENOMEM; + + skb_push(skb, length); + nh = (struct nshhdr *)(skb->data); + memcpy(nh, pushed_nh, length); + nh->np = next_proto; + skb_postpush_rcsum(skb, nh, length); + + skb->protocol = htons(ETH_P_NSH); + skb_reset_mac_header(skb); + skb_reset_network_header(skb); + skb_reset_mac_len(skb); + + return 0; +} +EXPORT_SYMBOL_GPL(nsh_push); + +int nsh_pop(struct sk_buff *skb) +{ + struct nshhdr *nh; + size_t length; + __be16 inner_proto; + + if (!pskb_may_pull(skb, NSH_BASE_HDR_LEN)) + return -ENOMEM; + nh = (struct nshhdr *)(skb->data); + length = nsh_hdr_len(nh); + inner_proto = tun_p_to_eth_p(nh->np); + if (!pskb_may_pull(skb, length)) + return -ENOMEM; + + if (!inner_proto) + return -EAFNOSUPPORT; + + skb_pull_rcsum(skb, length); + skb_reset_mac_header(skb); + skb_reset_network_header(skb); + skb_reset_mac_len(skb); + skb->protocol = inner_proto; + + return 0; +} +EXPORT_SYMBOL_GPL(nsh_pop); + static struct sk_buff *nsh_gso_segment(struct sk_buff *skb, netdev_features_t features) { diff --git a/net/openvswitch/Kconfig b/net/openvswitch/Kconfig index ce947292ae77..2650205cdaf9 100644 --- a/net/openvswitch/Kconfig +++ b/net/openvswitch/Kconfig @@ -14,6 +14,7 @@ config OPENVSWITCH select MPLS select NET_MPLS_GSO select DST_CACHE + select NET_NSH ---help--- Open vSwitch is a multilayer Ethernet switch targeted at virtualized environments. In addition to supporting a variety of features diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index a551232daf61..9a6a6d51e421 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -43,6 +43,7 @@ #include "flow.h" #include "conntrack.h" #include "vport.h" +#include "flow_netlink.h" struct deferred_action { struct sk_buff *skb; @@ -380,6 +381,38 @@ static int push_eth(struct sk_buff *skb, struct sw_flow_key *key, return 0; } +static int push_nsh(struct sk_buff *skb, struct sw_flow_key *key, + const struct nshhdr *nh) +{ + int err; + + err = nsh_push(skb, nh); + if (err) + return err; + + /* safe right before invalidate_flow_key */ + key->mac_proto = MAC_PROTO_NONE; + invalidate_flow_key(key); + return 0; +} + +static int pop_nsh(struct sk_buff *skb, struct sw_flow_key *key) +{ + int err; + + err = nsh_pop(skb); + if (err) + return err; + + /* safe right before invalidate_flow_key */ + if (skb->protocol == htons(ETH_P_TEB)) + key->mac_proto = MAC_PROTO_ETHERNET; + else + key->mac_proto = MAC_PROTO_NONE; + invalidate_flow_key(key); + return 0; +} + static void update_ip_l4_checksum(struct sk_buff *skb, struct iphdr *nh, __be32 addr, __be32 new_addr) { @@ -602,6 +635,69 @@ static int set_ipv6(struct sk_buff *skb, struct sw_flow_key *flow_key, return 0; } +static int set_nsh(struct sk_buff *skb, struct sw_flow_key *flow_key, + const struct nlattr *a) +{ + struct nshhdr *nh; + size_t length; + int err; + u8 flags; + u8 ttl; + int i; + + struct ovs_key_nsh key; + struct ovs_key_nsh mask; + + err = nsh_key_from_nlattr(a, &key, &mask); + if (err) + return err; + + /* Make sure the NSH base header is there */ + if (!pskb_may_pull(skb, skb_network_offset(skb) + NSH_BASE_HDR_LEN)) + return -ENOMEM; + + nh = nsh_hdr(skb); + length = nsh_hdr_len(nh); + + /* Make sure the whole NSH header is there */ + err = skb_ensure_writable(skb, skb_network_offset(skb) + + length); + if (unlikely(err)) + return err; + + nh = nsh_hdr(skb); + skb_postpull_rcsum(skb, nh, length); + flags = nsh_get_flags(nh); + flags = OVS_MASKED(flags, key.base.flags, mask.base.flags); + flow_key->nsh.base.flags = flags; + ttl = nsh_get_ttl(nh); + ttl = OVS_MASKED(ttl, key.base.ttl, mask.base.ttl); + flow_key->nsh.base.ttl = ttl; + nsh_set_flags_and_ttl(nh, flags, ttl); + nh->path_hdr = OVS_MASKED(nh->path_hdr, key.base.path_hdr, + mask.base.path_hdr); + flow_key->nsh.base.path_hdr = nh->path_hdr; + switch (nh->mdtype) { + case NSH_M_TYPE1: + for (i = 0; i < NSH_MD1_CONTEXT_SIZE; i++) { + nh->md1.context[i] = + OVS_MASKED(nh->md1.context[i], key.context[i], + mask.context[i]); + } + memcpy(flow_key->nsh.context, nh->md1.context, + sizeof(nh->md1.context)); + break; + case NSH_M_TYPE2: + memset(flow_key->nsh.context, 0, + sizeof(flow_key->nsh.context)); + break; + default: + return -EINVAL; + } + skb_postpush_rcsum(skb, nh, length); + return 0; +} + /* Must follow skb_ensure_writable() since that can move the skb data. */ static void set_tp_port(struct sk_buff *skb, __be16 *port, __be16 new_port, __sum16 *check) @@ -1024,6 +1120,10 @@ static int execute_masked_set_action(struct sk_buff *skb, get_mask(a, struct ovs_key_ethernet *)); break; + case OVS_KEY_ATTR_NSH: + err = set_nsh(skb, flow_key, a); + break; + case OVS_KEY_ATTR_IPV4: err = set_ipv4(skb, flow_key, nla_data(a), get_mask(a, struct ovs_key_ipv4 *)); @@ -1214,6 +1314,22 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb, case OVS_ACTION_ATTR_POP_ETH: err = pop_eth(skb, key); break; + + case OVS_ACTION_ATTR_PUSH_NSH: { + u8 buffer[NSH_HDR_MAX_LEN]; + struct nshhdr *nh = (struct nshhdr *)buffer; + + err = nsh_hdr_from_nlattr(nla_data(a), nh, + NSH_HDR_MAX_LEN); + if (unlikely(err)) + break; + err = push_nsh(skb, key, nh); + break; + } + + case OVS_ACTION_ATTR_POP_NSH: + err = pop_nsh(skb, key); + break; } if (unlikely(err)) { diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c index 8c94cef25a72..864ddb1e3642 100644 --- a/net/openvswitch/flow.c +++ b/net/openvswitch/flow.c @@ -46,6 +46,7 @@ #include #include #include +#include #include "conntrack.h" #include "datapath.h" @@ -490,6 +491,52 @@ invalid: return 0; } +static int parse_nsh(struct sk_buff *skb, struct sw_flow_key *key) +{ + struct nshhdr *nh; + unsigned int nh_ofs = skb_network_offset(skb); + u8 version, length; + int err; + + err = check_header(skb, nh_ofs + NSH_BASE_HDR_LEN); + if (unlikely(err)) + return err; + + nh = nsh_hdr(skb); + version = nsh_get_ver(nh); + length = nsh_hdr_len(nh); + + if (version != 0) + return -EINVAL; + + err = check_header(skb, nh_ofs + length); + if (unlikely(err)) + return err; + + nh = nsh_hdr(skb); + key->nsh.base.flags = nsh_get_flags(nh); + key->nsh.base.ttl = nsh_get_ttl(nh); + key->nsh.base.mdtype = nh->mdtype; + key->nsh.base.np = nh->np; + key->nsh.base.path_hdr = nh->path_hdr; + switch (key->nsh.base.mdtype) { + case NSH_M_TYPE1: + if (length != NSH_M_TYPE1_LEN) + return -EINVAL; + memcpy(key->nsh.context, nh->md1.context, + sizeof(nh->md1)); + break; + case NSH_M_TYPE2: + memset(key->nsh.context, 0, + sizeof(nh->md1)); + break; + default: + return -EINVAL; + } + + return 0; +} + /** * key_extract - extracts a flow key from an Ethernet frame. * @skb: sk_buff that contains the frame, with skb->data pointing to the @@ -735,6 +782,10 @@ static int key_extract(struct sk_buff *skb, struct sw_flow_key *key) memset(&key->tp, 0, sizeof(key->tp)); } } + } else if (key->eth.type == htons(ETH_P_NSH)) { + error = parse_nsh(skb, key); + if (error) + return error; } return 0; } diff --git a/net/openvswitch/flow.h b/net/openvswitch/flow.h index 1875bba4f865..c670dd24b8b7 100644 --- a/net/openvswitch/flow.h +++ b/net/openvswitch/flow.h @@ -35,6 +35,7 @@ #include #include #include +#include struct sk_buff; @@ -66,6 +67,11 @@ struct vlan_head { (offsetof(struct sw_flow_key, recirc_id) + \ FIELD_SIZEOF(struct sw_flow_key, recirc_id)) +struct ovs_key_nsh { + struct ovs_nsh_key_base base; + __be32 context[NSH_MD1_CONTEXT_SIZE]; +}; + struct sw_flow_key { u8 tun_opts[IP_TUNNEL_OPTS_MAX]; u8 tun_opts_len; @@ -143,6 +149,7 @@ struct sw_flow_key { } nd; }; } ipv6; + struct ovs_key_nsh nsh; /* network service header */ }; struct { /* Connection tracking fields not packed above. */ diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c index dc0d79092e74..4201f9293af3 100644 --- a/net/openvswitch/flow_netlink.c +++ b/net/openvswitch/flow_netlink.c @@ -48,6 +48,7 @@ #include #include #include +#include #include #include "flow_netlink.h" @@ -80,9 +81,11 @@ static bool actions_may_change_flow(const struct nlattr *actions) case OVS_ACTION_ATTR_HASH: case OVS_ACTION_ATTR_POP_ETH: case OVS_ACTION_ATTR_POP_MPLS: + case OVS_ACTION_ATTR_POP_NSH: case OVS_ACTION_ATTR_POP_VLAN: case OVS_ACTION_ATTR_PUSH_ETH: case OVS_ACTION_ATTR_PUSH_MPLS: + case OVS_ACTION_ATTR_PUSH_NSH: case OVS_ACTION_ATTR_PUSH_VLAN: case OVS_ACTION_ATTR_SAMPLE: case OVS_ACTION_ATTR_SET: @@ -175,7 +178,8 @@ static bool match_validate(const struct sw_flow_match *match, | (1 << OVS_KEY_ATTR_ICMPV6) | (1 << OVS_KEY_ATTR_ARP) | (1 << OVS_KEY_ATTR_ND) - | (1 << OVS_KEY_ATTR_MPLS)); + | (1 << OVS_KEY_ATTR_MPLS) + | (1 << OVS_KEY_ATTR_NSH)); /* Always allowed mask fields. */ mask_allowed |= ((1 << OVS_KEY_ATTR_TUNNEL) @@ -284,6 +288,14 @@ static bool match_validate(const struct sw_flow_match *match, } } + if (match->key->eth.type == htons(ETH_P_NSH)) { + key_expected |= 1 << OVS_KEY_ATTR_NSH; + if (match->mask && + match->mask->key.eth.type == htons(0xffff)) { + mask_allowed |= 1 << OVS_KEY_ATTR_NSH; + } + } + if ((key_attrs & key_expected) != key_expected) { /* Key attributes check failed. */ OVS_NLERR(log, "Missing key (keys=%llx, expected=%llx)", @@ -325,12 +337,25 @@ size_t ovs_tun_key_attr_size(void) + nla_total_size(4); /* OVS_TUNNEL_KEY_ATTR_ERSPAN_OPTS */ } +size_t ovs_nsh_key_attr_size(void) +{ + /* Whenever adding new OVS_NSH_KEY_ FIELDS, we should consider + * updating this function. + */ + return nla_total_size(NSH_BASE_HDR_LEN) /* OVS_NSH_KEY_ATTR_BASE */ + /* OVS_NSH_KEY_ATTR_MD1 and OVS_NSH_KEY_ATTR_MD2 are + * mutually exclusive, so the bigger one can cover + * the small one. + */ + + nla_total_size(NSH_CTX_HDRS_MAX_LEN); +} + size_t ovs_key_attr_size(void) { /* Whenever adding new OVS_KEY_ FIELDS, we should consider * updating this function. */ - BUILD_BUG_ON(OVS_KEY_ATTR_TUNNEL_INFO != 28); + BUILD_BUG_ON(OVS_KEY_ATTR_TUNNEL_INFO != 29); return nla_total_size(4) /* OVS_KEY_ATTR_PRIORITY */ + nla_total_size(0) /* OVS_KEY_ATTR_TUNNEL */ @@ -344,6 +369,8 @@ size_t ovs_key_attr_size(void) + nla_total_size(4) /* OVS_KEY_ATTR_CT_MARK */ + nla_total_size(16) /* OVS_KEY_ATTR_CT_LABELS */ + nla_total_size(40) /* OVS_KEY_ATTR_CT_ORIG_TUPLE_IPV6 */ + + nla_total_size(0) /* OVS_KEY_ATTR_NSH */ + + ovs_nsh_key_attr_size() + nla_total_size(12) /* OVS_KEY_ATTR_ETHERNET */ + nla_total_size(2) /* OVS_KEY_ATTR_ETHERTYPE */ + nla_total_size(4) /* OVS_KEY_ATTR_VLAN */ @@ -377,6 +404,13 @@ static const struct ovs_len_tbl ovs_tunnel_key_lens[OVS_TUNNEL_KEY_ATTR_MAX + 1] [OVS_TUNNEL_KEY_ATTR_ERSPAN_OPTS] = { .len = sizeof(u32) }, }; +static const struct ovs_len_tbl +ovs_nsh_key_attr_lens[OVS_NSH_KEY_ATTR_MAX + 1] = { + [OVS_NSH_KEY_ATTR_BASE] = { .len = sizeof(struct ovs_nsh_key_base) }, + [OVS_NSH_KEY_ATTR_MD1] = { .len = sizeof(struct ovs_nsh_key_md1) }, + [OVS_NSH_KEY_ATTR_MD2] = { .len = OVS_ATTR_VARIABLE }, +}; + /* The size of the argument for each %OVS_KEY_ATTR_* Netlink attribute. */ static const struct ovs_len_tbl ovs_key_lens[OVS_KEY_ATTR_MAX + 1] = { [OVS_KEY_ATTR_ENCAP] = { .len = OVS_ATTR_NESTED }, @@ -409,6 +443,8 @@ static const struct ovs_len_tbl ovs_key_lens[OVS_KEY_ATTR_MAX + 1] = { .len = sizeof(struct ovs_key_ct_tuple_ipv4) }, [OVS_KEY_ATTR_CT_ORIG_TUPLE_IPV6] = { .len = sizeof(struct ovs_key_ct_tuple_ipv6) }, + [OVS_KEY_ATTR_NSH] = { .len = OVS_ATTR_NESTED, + .next = ovs_nsh_key_attr_lens, }, }; static bool check_attr_len(unsigned int attr_len, unsigned int expected_len) @@ -1227,6 +1263,221 @@ static int metadata_from_nlattrs(struct net *net, struct sw_flow_match *match, return 0; } +int nsh_hdr_from_nlattr(const struct nlattr *attr, + struct nshhdr *nh, size_t size) +{ + struct nlattr *a; + int rem; + u8 flags = 0; + u8 ttl = 0; + int mdlen = 0; + + /* validate_nsh has check this, so we needn't do duplicate check here + */ + if (size < NSH_BASE_HDR_LEN) + return -ENOBUFS; + + nla_for_each_nested(a, attr, rem) { + int type = nla_type(a); + + switch (type) { + case OVS_NSH_KEY_ATTR_BASE: { + const struct ovs_nsh_key_base *base = nla_data(a); + + flags = base->flags; + ttl = base->ttl; + nh->np = base->np; + nh->mdtype = base->mdtype; + nh->path_hdr = base->path_hdr; + break; + } + case OVS_NSH_KEY_ATTR_MD1: + mdlen = nla_len(a); + if (mdlen > size - NSH_BASE_HDR_LEN) + return -ENOBUFS; + memcpy(&nh->md1, nla_data(a), mdlen); + break; + + case OVS_NSH_KEY_ATTR_MD2: + mdlen = nla_len(a); + if (mdlen > size - NSH_BASE_HDR_LEN) + return -ENOBUFS; + memcpy(&nh->md2, nla_data(a), mdlen); + break; + + default: + return -EINVAL; + } + } + + /* nsh header length = NSH_BASE_HDR_LEN + mdlen */ + nh->ver_flags_ttl_len = 0; + nsh_set_flags_ttl_len(nh, flags, ttl, NSH_BASE_HDR_LEN + mdlen); + + return 0; +} + +int nsh_key_from_nlattr(const struct nlattr *attr, + struct ovs_key_nsh *nsh, struct ovs_key_nsh *nsh_mask) +{ + struct nlattr *a; + int rem; + + /* validate_nsh has check this, so we needn't do duplicate check here + */ + nla_for_each_nested(a, attr, rem) { + int type = nla_type(a); + + switch (type) { + case OVS_NSH_KEY_ATTR_BASE: { + const struct ovs_nsh_key_base *base = nla_data(a); + const struct ovs_nsh_key_base *base_mask = base + 1; + + nsh->base = *base; + nsh_mask->base = *base_mask; + break; + } + case OVS_NSH_KEY_ATTR_MD1: { + const struct ovs_nsh_key_md1 *md1 = nla_data(a); + const struct ovs_nsh_key_md1 *md1_mask = md1 + 1; + + memcpy(nsh->context, md1->context, sizeof(*md1)); + memcpy(nsh_mask->context, md1_mask->context, + sizeof(*md1_mask)); + break; + } + case OVS_NSH_KEY_ATTR_MD2: + /* Not supported yet */ + return -ENOTSUPP; + default: + return -EINVAL; + } + } + + return 0; +} + +static int nsh_key_put_from_nlattr(const struct nlattr *attr, + struct sw_flow_match *match, bool is_mask, + bool is_push_nsh, bool log) +{ + struct nlattr *a; + int rem; + bool has_base = false; + bool has_md1 = false; + bool has_md2 = false; + u8 mdtype = 0; + int mdlen = 0; + + if (WARN_ON(is_push_nsh && is_mask)) + return -EINVAL; + + nla_for_each_nested(a, attr, rem) { + int type = nla_type(a); + int i; + + if (type > OVS_NSH_KEY_ATTR_MAX) { + OVS_NLERR(log, "nsh attr %d is out of range max %d", + type, OVS_NSH_KEY_ATTR_MAX); + return -EINVAL; + } + + if (!check_attr_len(nla_len(a), + ovs_nsh_key_attr_lens[type].len)) { + OVS_NLERR( + log, + "nsh attr %d has unexpected len %d expected %d", + type, + nla_len(a), + ovs_nsh_key_attr_lens[type].len + ); + return -EINVAL; + } + + switch (type) { + case OVS_NSH_KEY_ATTR_BASE: { + const struct ovs_nsh_key_base *base = nla_data(a); + + has_base = true; + mdtype = base->mdtype; + SW_FLOW_KEY_PUT(match, nsh.base.flags, + base->flags, is_mask); + SW_FLOW_KEY_PUT(match, nsh.base.ttl, + base->ttl, is_mask); + SW_FLOW_KEY_PUT(match, nsh.base.mdtype, + base->mdtype, is_mask); + SW_FLOW_KEY_PUT(match, nsh.base.np, + base->np, is_mask); + SW_FLOW_KEY_PUT(match, nsh.base.path_hdr, + base->path_hdr, is_mask); + break; + } + case OVS_NSH_KEY_ATTR_MD1: { + const struct ovs_nsh_key_md1 *md1 = nla_data(a); + + has_md1 = true; + for (i = 0; i < NSH_MD1_CONTEXT_SIZE; i++) + SW_FLOW_KEY_PUT(match, nsh.context[i], + md1->context[i], is_mask); + break; + } + case OVS_NSH_KEY_ATTR_MD2: + if (!is_push_nsh) /* Not supported MD type 2 yet */ + return -ENOTSUPP; + + has_md2 = true; + mdlen = nla_len(a); + if (mdlen > NSH_CTX_HDRS_MAX_LEN || mdlen <= 0) { + OVS_NLERR( + log, + "Invalid MD length %d for MD type %d", + mdlen, + mdtype + ); + return -EINVAL; + } + break; + default: + OVS_NLERR(log, "Unknown nsh attribute %d", + type); + return -EINVAL; + } + } + + if (rem > 0) { + OVS_NLERR(log, "nsh attribute has %d unknown bytes.", rem); + return -EINVAL; + } + + if (has_md1 && has_md2) { + OVS_NLERR( + 1, + "invalid nsh attribute: md1 and md2 are exclusive." + ); + return -EINVAL; + } + + if (!is_mask) { + if ((has_md1 && mdtype != NSH_M_TYPE1) || + (has_md2 && mdtype != NSH_M_TYPE2)) { + OVS_NLERR(1, "nsh attribute has unmatched MD type %d.", + mdtype); + return -EINVAL; + } + + if (is_push_nsh && + (!has_base || (!has_md1 && !has_md2))) { + OVS_NLERR( + 1, + "push_nsh: missing base or metadata attributes" + ); + return -EINVAL; + } + } + + return 0; +} + static int ovs_key_from_nlattrs(struct net *net, struct sw_flow_match *match, u64 attrs, const struct nlattr **a, bool is_mask, bool log) @@ -1354,6 +1605,13 @@ static int ovs_key_from_nlattrs(struct net *net, struct sw_flow_match *match, attrs &= ~(1 << OVS_KEY_ATTR_ARP); } + if (attrs & (1 << OVS_KEY_ATTR_NSH)) { + if (nsh_key_put_from_nlattr(a[OVS_KEY_ATTR_NSH], match, + is_mask, false, log) < 0) + return -EINVAL; + attrs &= ~(1 << OVS_KEY_ATTR_NSH); + } + if (attrs & (1 << OVS_KEY_ATTR_MPLS)) { const struct ovs_key_mpls *mpls_key; @@ -1670,6 +1928,34 @@ static int ovs_nla_put_vlan(struct sk_buff *skb, const struct vlan_head *vh, return 0; } +static int nsh_key_to_nlattr(const struct ovs_key_nsh *nsh, bool is_mask, + struct sk_buff *skb) +{ + struct nlattr *start; + + start = nla_nest_start(skb, OVS_KEY_ATTR_NSH); + if (!start) + return -EMSGSIZE; + + if (nla_put(skb, OVS_NSH_KEY_ATTR_BASE, sizeof(nsh->base), &nsh->base)) + goto nla_put_failure; + + if (is_mask || nsh->base.mdtype == NSH_M_TYPE1) { + if (nla_put(skb, OVS_NSH_KEY_ATTR_MD1, + sizeof(nsh->context), nsh->context)) + goto nla_put_failure; + } + + /* Don't support MD type 2 yet */ + + nla_nest_end(skb, start); + + return 0; + +nla_put_failure: + return -EMSGSIZE; +} + static int __ovs_nla_put_key(const struct sw_flow_key *swkey, const struct sw_flow_key *output, bool is_mask, struct sk_buff *skb) @@ -1798,6 +2084,9 @@ static int __ovs_nla_put_key(const struct sw_flow_key *swkey, ipv6_key->ipv6_tclass = output->ip.tos; ipv6_key->ipv6_hlimit = output->ip.ttl; ipv6_key->ipv6_frag = output->ip.frag; + } else if (swkey->eth.type == htons(ETH_P_NSH)) { + if (nsh_key_to_nlattr(&output->nsh, is_mask, skb)) + goto nla_put_failure; } else if (swkey->eth.type == htons(ETH_P_ARP) || swkey->eth.type == htons(ETH_P_RARP)) { struct ovs_key_arp *arp_key; @@ -2292,6 +2581,19 @@ static int validate_and_copy_set_tun(const struct nlattr *attr, return err; } +static bool validate_nsh(const struct nlattr *attr, bool is_mask, + bool is_push_nsh, bool log) +{ + struct sw_flow_match match; + struct sw_flow_key key; + int ret = 0; + + ovs_match_init(&match, &key, true, NULL); + ret = nsh_key_put_from_nlattr(attr, &match, is_mask, + is_push_nsh, log); + return !ret; +} + /* Return false if there are any non-masked bits set. * Mask follows data immediately, before any netlink padding. */ @@ -2434,6 +2736,13 @@ static int validate_set(const struct nlattr *a, break; + case OVS_KEY_ATTR_NSH: + if (eth_type != htons(ETH_P_NSH)) + return -EINVAL; + if (!validate_nsh(nla_data(a), masked, false, log)) + return -EINVAL; + break; + default: return -EINVAL; } @@ -2533,6 +2842,8 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr, [OVS_ACTION_ATTR_TRUNC] = sizeof(struct ovs_action_trunc), [OVS_ACTION_ATTR_PUSH_ETH] = sizeof(struct ovs_action_push_eth), [OVS_ACTION_ATTR_POP_ETH] = 0, + [OVS_ACTION_ATTR_PUSH_NSH] = (u32)-1, + [OVS_ACTION_ATTR_POP_NSH] = 0, }; const struct ovs_action_push_vlan *vlan; int type = nla_type(a); @@ -2690,6 +3001,34 @@ static int __ovs_nla_copy_actions(struct net *net, const struct nlattr *attr, mac_proto = MAC_PROTO_ETHERNET; break; + case OVS_ACTION_ATTR_PUSH_NSH: + if (mac_proto != MAC_PROTO_ETHERNET) { + u8 next_proto; + + next_proto = tun_p_from_eth_p(eth_type); + if (!next_proto) + return -EINVAL; + } + mac_proto = MAC_PROTO_NONE; + if (!validate_nsh(nla_data(a), false, true, true)) + return -EINVAL; + break; + + case OVS_ACTION_ATTR_POP_NSH: { + __be16 inner_proto; + + if (eth_type != htons(ETH_P_NSH)) + return -EINVAL; + inner_proto = tun_p_to_eth_p(key->nsh.base.np); + if (!inner_proto) + return -EINVAL; + if (key->nsh.base.np == TUN_P_ETHERNET) + mac_proto = MAC_PROTO_ETHERNET; + else + mac_proto = MAC_PROTO_NONE; + break; + } + default: OVS_NLERR(log, "Unknown Action type %d", type); return -EINVAL; diff --git a/net/openvswitch/flow_netlink.h b/net/openvswitch/flow_netlink.h index 929c665ac3aa..6657606b2b47 100644 --- a/net/openvswitch/flow_netlink.h +++ b/net/openvswitch/flow_netlink.h @@ -79,4 +79,9 @@ int ovs_nla_put_actions(const struct nlattr *attr, void ovs_nla_free_flow_actions(struct sw_flow_actions *); void ovs_nla_free_flow_actions_rcu(struct sw_flow_actions *); +int nsh_key_from_nlattr(const struct nlattr *attr, struct ovs_key_nsh *nsh, + struct ovs_key_nsh *nsh_mask); +int nsh_hdr_from_nlattr(const struct nlattr *attr, struct nshhdr *nh, + size_t size); + #endif /* flow_netlink.h */ -- cgit v1.2.3 From 24a9332a58b7f41a0d36c35a2c6897242bffdbc0 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 6 Nov 2017 16:11:43 -0500 Subject: net: dsa: constify cpu_dp member of dsa_port A DSA port has a dedicated CPU port assigned to it, stored in the cpu_dp member. It is not meant to be modified by a port, thus make it const. Signed-off-by: Vivien Didelot Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- include/net/dsa.h | 2 +- net/dsa/slave.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index e54332968417..2a8613b5a23d 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -190,7 +190,7 @@ struct dsa_port { struct dsa_switch *ds; unsigned int index; const char *name; - struct dsa_port *cpu_dp; + const struct dsa_port *cpu_dp; struct device_node *dn; unsigned int ageing_time; u8 stp_state; diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 814ced75a0cc..cc7fe47dd4bf 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -1147,7 +1147,7 @@ static void dsa_slave_notify(struct net_device *dev, unsigned long val) int dsa_slave_create(struct dsa_port *port) { - struct dsa_port *cpu_dp = port->cpu_dp; + const struct dsa_port *cpu_dp = port->cpu_dp; struct net_device *master = cpu_dp->master; struct dsa_switch *ds = port->ds; const char *name = port->name; -- cgit v1.2.3 From ec15dd4269d0cbf947c9a2dfdcf08a917098fab1 Mon Sep 17 00:00:00 2001 From: Vivien Didelot Date: Mon, 6 Nov 2017 16:11:46 -0500 Subject: net: dsa: setup and teardown tree This commit provides better scope for the DSA tree setup and teardown functions. It renames the "applied" bool to "setup" and print a message when the tree is setup, as it is done during teardown. At the same time, check dst->setup in dsa_tree_setup, where it is set to true. Signed-off-by: Vivien Didelot Signed-off-by: David S. Miller --- include/net/dsa.h | 2 +- net/dsa/dsa2.c | 32 ++++++++++++++++++-------------- 2 files changed, 19 insertions(+), 15 deletions(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 2a8613b5a23d..6c239257309b 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -122,7 +122,7 @@ struct dsa_switch_tree { struct kref refcount; /* Has this tree been applied to the hardware? */ - bool applied; + bool setup; /* * Configuration data for the platform device that owns diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index c9b50339fcac..1a8df0a177b5 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -486,12 +486,18 @@ static void dsa_tree_teardown_master(struct dsa_switch_tree *dst) return dsa_master_teardown(master); } -static int dsa_dst_apply(struct dsa_switch_tree *dst) +static int dsa_tree_setup(struct dsa_switch_tree *dst) { struct dsa_switch *ds; u32 index; int err; + if (dst->setup) { + pr_err("DSA: tree %d already setup! Disjoint trees?\n", + dst->index); + return -EEXIST; + } + err = dsa_tree_setup_default_cpu(dst); if (err) return err; @@ -510,17 +516,19 @@ static int dsa_dst_apply(struct dsa_switch_tree *dst) if (err) return err; - dst->applied = true; + dst->setup = true; + + pr_info("DSA: tree %d setup\n", dst->index); return 0; } -static void dsa_dst_unapply(struct dsa_switch_tree *dst) +static void dsa_tree_teardown(struct dsa_switch_tree *dst) { struct dsa_switch *ds; u32 index; - if (!dst->applied) + if (!dst->setup) return; dsa_tree_teardown_master(dst); @@ -535,8 +543,9 @@ static void dsa_dst_unapply(struct dsa_switch_tree *dst) dsa_tree_teardown_default_cpu(dst); - pr_info("DSA: tree %d unapplied\n", dst->index); - dst->applied = false; + pr_info("DSA: tree %d torn down\n", dst->index); + + dst->setup = false; } static void dsa_tree_remove_switch(struct dsa_switch_tree *dst, @@ -794,14 +803,9 @@ static int _dsa_register_switch(struct dsa_switch *ds) if (err == 1) return 0; - if (dst->applied) { - pr_info("DSA: Disjoint trees?\n"); - return -EINVAL; - } - - err = dsa_dst_apply(dst); + err = dsa_tree_setup(dst); if (err) { - dsa_dst_unapply(dst); + dsa_tree_teardown(dst); goto out_del_dst; } @@ -852,7 +856,7 @@ static void _dsa_unregister_switch(struct dsa_switch *ds) struct dsa_switch_tree *dst = ds->dst; unsigned int index = ds->index; - dsa_dst_unapply(dst); + dsa_tree_teardown(dst); dsa_tree_remove_switch(dst, index); } -- cgit v1.2.3 From 47d5b6db2afa766d7af85db684d0b5f092e4fc46 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Thu, 9 Nov 2017 23:10:59 +0100 Subject: net: bridge: Add/del switchdev object on host join/leave When the host joins or leaves a multicast group, use switchdev to add an object to the hardware to forward traffic for the group to the host. Signed-off-by: Andrew Lunn Acked-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- include/net/switchdev.h | 1 + net/bridge/br_mdb.c | 43 +++++++++++++++++++++++++++++++++++++++++++ net/switchdev/switchdev.c | 2 ++ 3 files changed, 46 insertions(+) (limited to 'include/net') diff --git a/include/net/switchdev.h b/include/net/switchdev.h index d756fbe46625..39bc855d7fee 100644 --- a/include/net/switchdev.h +++ b/include/net/switchdev.h @@ -76,6 +76,7 @@ enum switchdev_obj_id { SWITCHDEV_OBJ_ID_UNDEFINED, SWITCHDEV_OBJ_ID_PORT_VLAN, SWITCHDEV_OBJ_ID_PORT_MDB, + SWITCHDEV_OBJ_ID_HOST_MDB, }; struct switchdev_obj { diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c index 702408d2a93c..b0f4c734900b 100644 --- a/net/bridge/br_mdb.c +++ b/net/bridge/br_mdb.c @@ -292,6 +292,46 @@ err: kfree(priv); } +static void br_mdb_switchdev_host_port(struct net_device *dev, + struct net_device *lower_dev, + struct br_mdb_entry *entry, int type) +{ + struct switchdev_obj_port_mdb mdb = { + .obj = { + .id = SWITCHDEV_OBJ_ID_HOST_MDB, + .flags = SWITCHDEV_F_DEFER, + }, + .vid = entry->vid, + }; + + if (entry->addr.proto == htons(ETH_P_IP)) + ip_eth_mc_map(entry->addr.u.ip4, mdb.addr); +#if IS_ENABLED(CONFIG_IPV6) + else + ipv6_eth_mc_map(&entry->addr.u.ip6, mdb.addr); +#endif + + mdb.obj.orig_dev = dev; + switch (type) { + case RTM_NEWMDB: + switchdev_port_obj_add(lower_dev, &mdb.obj); + break; + case RTM_DELMDB: + switchdev_port_obj_del(lower_dev, &mdb.obj); + break; + } +} + +static void br_mdb_switchdev_host(struct net_device *dev, + struct br_mdb_entry *entry, int type) +{ + struct net_device *lower_dev; + struct list_head *iter; + + netdev_for_each_lower_dev(dev, lower_dev, iter) + br_mdb_switchdev_host_port(dev, lower_dev, entry, type); +} + static void __br_mdb_notify(struct net_device *dev, struct net_bridge_port *p, struct br_mdb_entry *entry, int type) { @@ -331,6 +371,9 @@ static void __br_mdb_notify(struct net_device *dev, struct net_bridge_port *p, switchdev_port_obj_del(port_dev, &mdb.obj); } + if (!p) + br_mdb_switchdev_host(dev, entry, type); + skb = nlmsg_new(rtnl_mdb_nlmsg_size(), GFP_ATOMIC); if (!skb) goto errout; diff --git a/net/switchdev/switchdev.c b/net/switchdev/switchdev.c index 0531b41d1f2d..74b9d916a58b 100644 --- a/net/switchdev/switchdev.c +++ b/net/switchdev/switchdev.c @@ -345,6 +345,8 @@ static size_t switchdev_obj_size(const struct switchdev_obj *obj) return sizeof(struct switchdev_obj_port_vlan); case SWITCHDEV_OBJ_ID_PORT_MDB: return sizeof(struct switchdev_obj_port_mdb); + case SWITCHDEV_OBJ_ID_HOST_MDB: + return sizeof(struct switchdev_obj_port_mdb); default: BUG(); } -- cgit v1.2.3 From a3dcaf17ee54f1d01d22cc2b22cab0b4f60d78cf Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 7 Nov 2017 00:29:27 -0800 Subject: net: allow per netns sysctl_rmem and sysctl_wmem for protos As we want to gradually implement per netns sysctl_rmem and sysctl_wmem on per protocol basis, add two new fields in struct proto, and two new helpers : sk_get_wmem0() and sk_get_rmem0() First user will be TCP. Then UDP and SCTP can be easily converted, while DECNET probably wont get this support. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/sock.h | 22 ++++++++++++++++++++++ include/trace/events/sock.h | 2 +- net/core/sock.c | 10 ++++++---- 3 files changed, 29 insertions(+), 5 deletions(-) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index 6f1be9726e02..688a823dccc3 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1101,8 +1101,12 @@ struct proto { */ unsigned long *memory_pressure; long *sysctl_mem; + int *sysctl_wmem; int *sysctl_rmem; + u32 sysctl_wmem_offset; + u32 sysctl_rmem_offset; + int max_header; bool no_autobind; @@ -2390,4 +2394,22 @@ extern int sysctl_optmem_max; extern __u32 sysctl_wmem_default; extern __u32 sysctl_rmem_default; +static inline int sk_get_wmem0(const struct sock *sk, const struct proto *proto) +{ + /* Does this proto have per netns sysctl_wmem ? */ + if (proto->sysctl_wmem_offset) + return *(int *)((void *)sock_net(sk) + proto->sysctl_wmem_offset); + + return *proto->sysctl_wmem; +} + +static inline int sk_get_rmem0(const struct sock *sk, const struct proto *proto) +{ + /* Does this proto have per netns sysctl_rmem ? */ + if (proto->sysctl_rmem_offset) + return *(int *)((void *)sock_net(sk) + proto->sysctl_rmem_offset); + + return *proto->sysctl_rmem; +} + #endif /* _SOCK_H */ diff --git a/include/trace/events/sock.h b/include/trace/events/sock.h index 6d31c0520ef3..ec4dade24466 100644 --- a/include/trace/events/sock.h +++ b/include/trace/events/sock.h @@ -48,7 +48,7 @@ TRACE_EVENT(sock_exceed_buf_limit, strncpy(__entry->name, prot->name, 32); __entry->sysctl_mem = prot->sysctl_mem; __entry->allocated = allocated; - __entry->sysctl_rmem = prot->sysctl_rmem[0]; + __entry->sysctl_rmem = sk_get_rmem0(sk, prot); __entry->rmem_alloc = atomic_read(&sk->sk_rmem_alloc); ), diff --git a/net/core/sock.c b/net/core/sock.c index 759400053110..c59bcf90d905 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -2346,16 +2346,18 @@ int __sk_mem_raise_allocated(struct sock *sk, int size, int amt, int kind) /* guarantee minimum buffer size under pressure */ if (kind == SK_MEM_RECV) { - if (atomic_read(&sk->sk_rmem_alloc) < prot->sysctl_rmem[0]) + if (atomic_read(&sk->sk_rmem_alloc) < sk_get_rmem0(sk, prot)) return 1; } else { /* SK_MEM_SEND */ + int wmem0 = sk_get_wmem0(sk, prot); + if (sk->sk_type == SOCK_STREAM) { - if (sk->sk_wmem_queued < prot->sysctl_wmem[0]) + if (sk->sk_wmem_queued < wmem0) return 1; - } else if (refcount_read(&sk->sk_wmem_alloc) < - prot->sysctl_wmem[0]) + } else if (refcount_read(&sk->sk_wmem_alloc) < wmem0) { return 1; + } } if (sk_has_memory_pressure(sk)) { -- cgit v1.2.3 From 356d1833b638bd465672aefeb71def3ab93fc17d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 7 Nov 2017 00:29:28 -0800 Subject: tcp: Namespace-ify sysctl_tcp_rmem and sysctl_tcp_wmem Note that when a new netns is created, it inherits its sysctl_tcp_rmem and sysctl_tcp_wmem from initial netns. This change is needed so that we can refine TCP rcvbuf autotuning, to take RTT into consideration. Signed-off-by: Eric Dumazet Cc: Wei Wang Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 2 ++ include/net/tcp.h | 2 -- net/ipv4/sysctl_net_ipv4.c | 32 ++++++++++++++++---------------- net/ipv4/tcp.c | 21 ++++++++------------- net/ipv4/tcp_input.c | 14 ++++++++------ net/ipv4/tcp_ipv4.c | 13 ++++++++++--- net/ipv4/tcp_output.c | 2 +- net/ipv6/tcp_ipv6.c | 4 ++-- 8 files changed, 47 insertions(+), 43 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 379550f8124a..5e12975fc658 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -155,6 +155,8 @@ struct netns_ipv4 { int sysctl_tcp_invalid_ratelimit; int sysctl_tcp_pacing_ss_ratio; int sysctl_tcp_pacing_ca_ratio; + int sysctl_tcp_wmem[3]; + int sysctl_tcp_rmem[3]; struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; diff --git a/include/net/tcp.h b/include/net/tcp.h index babfd4da1515..2f2c69ad31b2 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -242,8 +242,6 @@ void tcp_time_wait(struct sock *sk, int state, int timeo); /* sysctl variables for tcp */ extern int sysctl_tcp_max_orphans; extern long sysctl_tcp_mem[3]; -extern int sysctl_tcp_wmem[3]; -extern int sysctl_tcp_rmem[3]; #define TCP_RACK_LOSS_DETECTION 0x1 /* Use RACK to detect losses */ #define TCP_RACK_STATIC_REO_WND 0x2 /* Use static RACK reo wnd */ diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index a82b44038308..ef0ff3357a44 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -440,22 +440,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_doulongvec_minmax, }, - { - .procname = "tcp_wmem", - .data = &sysctl_tcp_wmem, - .maxlen = sizeof(sysctl_tcp_wmem), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &one, - }, - { - .procname = "tcp_rmem", - .data = &sysctl_tcp_rmem, - .maxlen = sizeof(sysctl_tcp_rmem), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &one, - }, { .procname = "tcp_low_latency", .data = &sysctl_tcp_low_latency, @@ -1164,6 +1148,22 @@ static struct ctl_table ipv4_net_table[] = { .extra1 = &zero, .extra2 = &thousand, }, + { + .procname = "tcp_wmem", + .data = &init_net.ipv4.sysctl_tcp_wmem, + .maxlen = sizeof(init_net.ipv4.sysctl_tcp_wmem), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &one, + }, + { + .procname = "tcp_rmem", + .data = &init_net.ipv4.sysctl_tcp_rmem, + .maxlen = sizeof(init_net.ipv4.sysctl_tcp_rmem), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &one, + }, { } }; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index c4cb19ed4628..bc71a27d5ad9 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -289,12 +289,7 @@ struct percpu_counter tcp_orphan_count; EXPORT_SYMBOL_GPL(tcp_orphan_count); long sysctl_tcp_mem[3] __read_mostly; -int sysctl_tcp_wmem[3] __read_mostly; -int sysctl_tcp_rmem[3] __read_mostly; - EXPORT_SYMBOL(sysctl_tcp_mem); -EXPORT_SYMBOL(sysctl_tcp_rmem); -EXPORT_SYMBOL(sysctl_tcp_wmem); atomic_long_t tcp_memory_allocated; /* Current allocated memory. */ EXPORT_SYMBOL(tcp_memory_allocated); @@ -456,8 +451,8 @@ void tcp_init_sock(struct sock *sk) icsk->icsk_sync_mss = tcp_sync_mss; - sk->sk_sndbuf = sysctl_tcp_wmem[1]; - sk->sk_rcvbuf = sysctl_tcp_rmem[1]; + sk->sk_sndbuf = sock_net(sk)->ipv4.sysctl_tcp_wmem[1]; + sk->sk_rcvbuf = sock_net(sk)->ipv4.sysctl_tcp_rmem[1]; sk_sockets_allocated_inc(sk); } @@ -3636,13 +3631,13 @@ void __init tcp_init(void) max_wshare = min(4UL*1024*1024, limit); max_rshare = min(6UL*1024*1024, limit); - sysctl_tcp_wmem[0] = SK_MEM_QUANTUM; - sysctl_tcp_wmem[1] = 16*1024; - sysctl_tcp_wmem[2] = max(64*1024, max_wshare); + init_net.ipv4.sysctl_tcp_wmem[0] = SK_MEM_QUANTUM; + init_net.ipv4.sysctl_tcp_wmem[1] = 16*1024; + init_net.ipv4.sysctl_tcp_wmem[2] = max(64*1024, max_wshare); - sysctl_tcp_rmem[0] = SK_MEM_QUANTUM; - sysctl_tcp_rmem[1] = 87380; - sysctl_tcp_rmem[2] = max(87380, max_rshare); + init_net.ipv4.sysctl_tcp_rmem[0] = SK_MEM_QUANTUM; + init_net.ipv4.sysctl_tcp_rmem[1] = 87380; + init_net.ipv4.sysctl_tcp_rmem[2] = max(87380, max_rshare); pr_info("Hash tables configured (established %u bind %u)\n", tcp_hashinfo.ehash_mask + 1, tcp_hashinfo.bhash_size); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index b54ee09cbcf7..9ceaa1fdc3ab 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -320,7 +320,7 @@ static void tcp_sndbuf_expand(struct sock *sk) sndmem *= nr_segs * per_mss; if (sk->sk_sndbuf < sndmem) - sk->sk_sndbuf = min(sndmem, sysctl_tcp_wmem[2]); + sk->sk_sndbuf = min(sndmem, sock_net(sk)->ipv4.sysctl_tcp_wmem[2]); } /* 2. Tuning advertised window (window_clamp, rcv_ssthresh) @@ -354,7 +354,7 @@ static int __tcp_grow_window(const struct sock *sk, const struct sk_buff *skb) struct tcp_sock *tp = tcp_sk(sk); /* Optimize this! */ int truesize = tcp_win_from_space(sk, skb->truesize) >> 1; - int window = tcp_win_from_space(sk, sysctl_tcp_rmem[2]) >> 1; + int window = tcp_win_from_space(sk, sock_net(sk)->ipv4.sysctl_tcp_rmem[2]) >> 1; while (tp->rcv_ssthresh <= window) { if (truesize <= skb->len) @@ -409,7 +409,7 @@ static void tcp_fixup_rcvbuf(struct sock *sk) rcvmem <<= 2; if (sk->sk_rcvbuf < rcvmem) - sk->sk_rcvbuf = min(rcvmem, sysctl_tcp_rmem[2]); + sk->sk_rcvbuf = min(rcvmem, sock_net(sk)->ipv4.sysctl_tcp_rmem[2]); } /* 4. Try to fixup all. It is made immediately after connection enters @@ -457,15 +457,16 @@ static void tcp_clamp_window(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); struct inet_connection_sock *icsk = inet_csk(sk); + struct net *net = sock_net(sk); icsk->icsk_ack.quick = 0; - if (sk->sk_rcvbuf < sysctl_tcp_rmem[2] && + if (sk->sk_rcvbuf < net->ipv4.sysctl_tcp_rmem[2] && !(sk->sk_userlocks & SOCK_RCVBUF_LOCK) && !tcp_under_memory_pressure(sk) && sk_memory_allocated(sk) < sk_prot_mem_limits(sk, 0)) { sk->sk_rcvbuf = min(atomic_read(&sk->sk_rmem_alloc), - sysctl_tcp_rmem[2]); + net->ipv4.sysctl_tcp_rmem[2]); } if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf) tp->rcv_ssthresh = min(tp->window_clamp, 2U * tp->advmss); @@ -623,7 +624,8 @@ void tcp_rcv_space_adjust(struct sock *sk) while (tcp_win_from_space(sk, rcvmem) < tp->advmss) rcvmem += 128; - rcvbuf = min(rcvwin / tp->advmss * rcvmem, sysctl_tcp_rmem[2]); + rcvbuf = min(rcvwin / tp->advmss * rcvmem, + sock_net(sk)->ipv4.sysctl_tcp_rmem[2]); if (rcvbuf > sk->sk_rcvbuf) { sk->sk_rcvbuf = rcvbuf; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 0162c577bb9c..1eac84b8044e 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2409,8 +2409,8 @@ struct proto tcp_prot = { .memory_allocated = &tcp_memory_allocated, .memory_pressure = &tcp_memory_pressure, .sysctl_mem = sysctl_tcp_mem, - .sysctl_wmem = sysctl_tcp_wmem, - .sysctl_rmem = sysctl_tcp_rmem, + .sysctl_wmem_offset = offsetof(struct net, ipv4.sysctl_tcp_wmem), + .sysctl_rmem_offset = offsetof(struct net, ipv4.sysctl_tcp_rmem), .max_header = MAX_TCP_HEADER, .obj_size = sizeof(struct tcp_sock), .slab_flags = SLAB_TYPESAFE_BY_RCU, @@ -2509,7 +2509,14 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_invalid_ratelimit = HZ/2; net->ipv4.sysctl_tcp_pacing_ss_ratio = 200; net->ipv4.sysctl_tcp_pacing_ca_ratio = 120; - + if (net != &init_net) { + memcpy(net->ipv4.sysctl_tcp_rmem, + init_net.ipv4.sysctl_tcp_rmem, + sizeof(init_net.ipv4.sysctl_tcp_rmem)); + memcpy(net->ipv4.sysctl_tcp_wmem, + init_net.ipv4.sysctl_tcp_wmem, + sizeof(init_net.ipv4.sysctl_tcp_wmem)); + } net->ipv4.sysctl_tcp_fastopen = TFO_CLIENT_ENABLE; spin_lock_init(&net->ipv4.tcp_fastopen_ctx_lock); net->ipv4.sysctl_tcp_fastopen_blackhole_timeout = 60 * 60; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index a9d917e4dad5..9b98d35aa0d8 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -220,7 +220,7 @@ void tcp_select_initial_window(const struct sock *sk, int __space, __u32 mss, (*rcv_wscale) = 0; if (wscale_ok) { /* Set window scaling on max possible window */ - space = max_t(u32, space, sysctl_tcp_rmem[2]); + space = max_t(u32, space, sock_net(sk)->ipv4.sysctl_tcp_rmem[2]); space = max_t(u32, space, sysctl_rmem_max); space = min_t(u32, space, *window_clamp); while (space > U16_MAX && (*rcv_wscale) < TCP_MAX_WSCALE) { diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 0e2529958b52..6bb98c93edfe 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1940,8 +1940,8 @@ struct proto tcpv6_prot = { .memory_pressure = &tcp_memory_pressure, .orphan_count = &tcp_orphan_count, .sysctl_mem = sysctl_tcp_mem, - .sysctl_wmem = sysctl_tcp_wmem, - .sysctl_rmem = sysctl_tcp_rmem, + .sysctl_wmem_offset = offsetof(struct net, ipv4.sysctl_tcp_wmem), + .sysctl_rmem_offset = offsetof(struct net, ipv4.sysctl_tcp_rmem), .max_header = MAX_TCP_HEADER, .obj_size = sizeof(struct tcp6_sock), .slab_flags = SLAB_TYPESAFE_BY_RCU, -- cgit v1.2.3 From 4c5b9d9642c859f7369338fc42c0f62f4151bef3 Mon Sep 17 00:00:00 2001 From: Manish Kurup Date: Tue, 7 Nov 2017 15:49:05 -0500 Subject: act_vlan: VLAN action rewrite to use RCU lock/unlock and update Using a spinlock in the VLAN action causes performance issues when the VLAN action is used on multiple cores. Rewrote the VLAN action to use RCU read locking for reads and updates instead. All functions now use an RCU dereferenced pointer to access the VLAN action context. Modified helper functions used by other modules, to use the RCU as opposed to directly accessing the structure. Acked-by: Jamal Hadi Salim Acked-by: Jiri Pirko Signed-off-by: Manish Kurup Signed-off-by: David S. Miller --- include/net/tc_act/tc_vlan.h | 46 +++++++++++++++++++++------ net/sched/act_vlan.c | 75 ++++++++++++++++++++++++++++++-------------- 2 files changed, 88 insertions(+), 33 deletions(-) (limited to 'include/net') diff --git a/include/net/tc_act/tc_vlan.h b/include/net/tc_act/tc_vlan.h index c2090df944ff..22ae260d6869 100644 --- a/include/net/tc_act/tc_vlan.h +++ b/include/net/tc_act/tc_vlan.h @@ -13,12 +13,17 @@ #include #include +struct tcf_vlan_params { + int tcfv_action; + u16 tcfv_push_vid; + __be16 tcfv_push_proto; + u8 tcfv_push_prio; + struct rcu_head rcu; +}; + struct tcf_vlan { struct tc_action common; - int tcfv_action; - u16 tcfv_push_vid; - __be16 tcfv_push_proto; - u8 tcfv_push_prio; + struct tcf_vlan_params __rcu *vlan_p; }; #define to_vlan(a) ((struct tcf_vlan *)a) @@ -33,22 +38,45 @@ static inline bool is_tcf_vlan(const struct tc_action *a) static inline u32 tcf_vlan_action(const struct tc_action *a) { - return to_vlan(a)->tcfv_action; + u32 tcfv_action; + + rcu_read_lock(); + tcfv_action = rcu_dereference(to_vlan(a)->vlan_p)->tcfv_action; + rcu_read_unlock(); + + return tcfv_action; } static inline u16 tcf_vlan_push_vid(const struct tc_action *a) { - return to_vlan(a)->tcfv_push_vid; + u16 tcfv_push_vid; + + rcu_read_lock(); + tcfv_push_vid = rcu_dereference(to_vlan(a)->vlan_p)->tcfv_push_vid; + rcu_read_unlock(); + + return tcfv_push_vid; } static inline __be16 tcf_vlan_push_proto(const struct tc_action *a) { - return to_vlan(a)->tcfv_push_proto; + __be16 tcfv_push_proto; + + rcu_read_lock(); + tcfv_push_proto = rcu_dereference(to_vlan(a)->vlan_p)->tcfv_push_proto; + rcu_read_unlock(); + + return tcfv_push_proto; } static inline u8 tcf_vlan_push_prio(const struct tc_action *a) { - return to_vlan(a)->tcfv_push_prio; -} + u8 tcfv_push_prio; + rcu_read_lock(); + tcfv_push_prio = rcu_dereference(to_vlan(a)->vlan_p)->tcfv_push_prio; + rcu_read_unlock(); + + return tcfv_push_prio; +} #endif /* __NET_TC_VLAN_H */ diff --git a/net/sched/act_vlan.c b/net/sched/act_vlan.c index b093badc1450..97f717a13ad5 100644 --- a/net/sched/act_vlan.c +++ b/net/sched/act_vlan.c @@ -26,6 +26,7 @@ static int tcf_vlan(struct sk_buff *skb, const struct tc_action *a, struct tcf_result *res) { struct tcf_vlan *v = to_vlan(a); + struct tcf_vlan_params *p; int action; int err; u16 tci; @@ -33,24 +34,27 @@ static int tcf_vlan(struct sk_buff *skb, const struct tc_action *a, tcf_lastuse_update(&v->tcf_tm); bstats_cpu_update(this_cpu_ptr(v->common.cpu_bstats), skb); - spin_lock(&v->tcf_lock); - action = v->tcf_action; - /* Ensure 'data' points at mac_header prior calling vlan manipulating * functions. */ if (skb_at_tc_ingress(skb)) skb_push_rcsum(skb, skb->mac_len); - switch (v->tcfv_action) { + rcu_read_lock(); + + action = READ_ONCE(v->tcf_action); + + p = rcu_dereference(v->vlan_p); + + switch (p->tcfv_action) { case TCA_VLAN_ACT_POP: err = skb_vlan_pop(skb); if (err) goto drop; break; case TCA_VLAN_ACT_PUSH: - err = skb_vlan_push(skb, v->tcfv_push_proto, v->tcfv_push_vid | - (v->tcfv_push_prio << VLAN_PRIO_SHIFT)); + err = skb_vlan_push(skb, p->tcfv_push_proto, p->tcfv_push_vid | + (p->tcfv_push_prio << VLAN_PRIO_SHIFT)); if (err) goto drop; break; @@ -69,14 +73,14 @@ static int tcf_vlan(struct sk_buff *skb, const struct tc_action *a, goto drop; } /* replace the vid */ - tci = (tci & ~VLAN_VID_MASK) | v->tcfv_push_vid; + tci = (tci & ~VLAN_VID_MASK) | p->tcfv_push_vid; /* replace prio bits, if tcfv_push_prio specified */ - if (v->tcfv_push_prio) { + if (p->tcfv_push_prio) { tci &= ~VLAN_PRIO_MASK; - tci |= v->tcfv_push_prio << VLAN_PRIO_SHIFT; + tci |= p->tcfv_push_prio << VLAN_PRIO_SHIFT; } /* put updated tci as hwaccel tag */ - __vlan_hwaccel_put_tag(skb, v->tcfv_push_proto, tci); + __vlan_hwaccel_put_tag(skb, p->tcfv_push_proto, tci); break; default: BUG(); @@ -89,10 +93,10 @@ drop: qstats_drop_inc(this_cpu_ptr(v->common.cpu_qstats)); unlock: + rcu_read_unlock(); if (skb_at_tc_ingress(skb)) skb_pull_rcsum(skb, skb->mac_len); - spin_unlock(&v->tcf_lock); return action; } @@ -109,6 +113,7 @@ static int tcf_vlan_init(struct net *net, struct nlattr *nla, { struct tc_action_net *tn = net_generic(net, vlan_net_id); struct nlattr *tb[TCA_VLAN_MAX + 1]; + struct tcf_vlan_params *p, *p_old; struct tc_vlan *parm; struct tcf_vlan *v; int action; @@ -187,46 +192,67 @@ static int tcf_vlan_init(struct net *net, struct nlattr *nla, v = to_vlan(*a); - spin_lock_bh(&v->tcf_lock); - - v->tcfv_action = action; - v->tcfv_push_vid = push_vid; - v->tcfv_push_prio = push_prio; - v->tcfv_push_proto = push_proto; + ASSERT_RTNL(); + p = kzalloc(sizeof(*p), GFP_KERNEL); + if (!p) { + if (ovr) + tcf_idr_release(*a, bind); + return -ENOMEM; + } v->tcf_action = parm->action; - spin_unlock_bh(&v->tcf_lock); + p_old = rtnl_dereference(v->vlan_p); + + p->tcfv_action = action; + p->tcfv_push_vid = push_vid; + p->tcfv_push_prio = push_prio; + p->tcfv_push_proto = push_proto; + + rcu_assign_pointer(v->vlan_p, p); + + if (p_old) + kfree_rcu(p_old, rcu); if (ret == ACT_P_CREATED) tcf_idr_insert(tn, *a); return ret; } +static void tcf_vlan_cleanup(struct tc_action *a, int bind) +{ + struct tcf_vlan *v = to_vlan(a); + struct tcf_vlan_params *p; + + p = rcu_dereference_protected(v->vlan_p, 1); + kfree_rcu(p, rcu); +} + static int tcf_vlan_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); struct tcf_vlan *v = to_vlan(a); + struct tcf_vlan_params *p = rtnl_dereference(v->vlan_p); struct tc_vlan opt = { .index = v->tcf_index, .refcnt = v->tcf_refcnt - ref, .bindcnt = v->tcf_bindcnt - bind, .action = v->tcf_action, - .v_action = v->tcfv_action, + .v_action = p->tcfv_action, }; struct tcf_t t; if (nla_put(skb, TCA_VLAN_PARMS, sizeof(opt), &opt)) goto nla_put_failure; - if ((v->tcfv_action == TCA_VLAN_ACT_PUSH || - v->tcfv_action == TCA_VLAN_ACT_MODIFY) && - (nla_put_u16(skb, TCA_VLAN_PUSH_VLAN_ID, v->tcfv_push_vid) || + if ((p->tcfv_action == TCA_VLAN_ACT_PUSH || + p->tcfv_action == TCA_VLAN_ACT_MODIFY) && + (nla_put_u16(skb, TCA_VLAN_PUSH_VLAN_ID, p->tcfv_push_vid) || nla_put_be16(skb, TCA_VLAN_PUSH_VLAN_PROTOCOL, - v->tcfv_push_proto) || + p->tcfv_push_proto) || (nla_put_u8(skb, TCA_VLAN_PUSH_VLAN_PRIORITY, - v->tcfv_push_prio)))) + p->tcfv_push_prio)))) goto nla_put_failure; tcf_tm_dump(&t, &v->tcf_tm); @@ -262,6 +288,7 @@ static struct tc_action_ops act_vlan_ops = { .act = tcf_vlan, .dump = tcf_vlan_dump, .init = tcf_vlan_init, + .cleanup = tcf_vlan_cleanup, .walk = tcf_vlan_walker, .lookup = tcf_vlan_search, .size = sizeof(struct tcf_vlan), -- cgit v1.2.3 From 8d6e79d3ce13e34957de87f7584cbf1bcde74c57 Mon Sep 17 00:00:00 2001 From: Jon Maloy Date: Wed, 8 Nov 2017 09:59:26 +0100 Subject: tipc: improve link resiliency when rps is activated Currently, the TIPC RPS dissector is based only on the incoming packets' source node address, hence steering all traffic from a node to the same core. We have seen that this makes the links vulnerable to starvation and unnecessary resets when we turn down the link tolerance to very low values. To reduce the risk of this happening, we exempt probe and probe replies packets from the convergence to one core per source node. Instead, we do the opposite, - we try to diverge those packets across as many cores as possible, by randomizing the flow selector key. To make such packets identifiable to the dissector, we add a new 'is_keepalive' bit to word 0 of the LINK_PROTOCOL header. This bit is set both for PROBE and PROBE_REPLY messages, and only for those. It should be noted that these packets are not part of any flow anyway, and only constitute a minuscule fraction of all packets sent across a link. Hence, there is no risk that this will affect overall performance. Acked-by: Ying Xue Signed-off-by: Jon Maloy Signed-off-by: David S. Miller --- include/net/flow_dissector.h | 12 ++++----- include/net/tipc.h | 62 ++++++++++++++++++++++++++++++++++++++++++++ net/core/flow_dissector.c | 30 ++++++++++----------- net/tipc/link.c | 26 +++++++++++-------- net/tipc/msg.h | 10 +++++++ 5 files changed, 108 insertions(+), 32 deletions(-) create mode 100644 include/net/tipc.h (limited to 'include/net') diff --git a/include/net/flow_dissector.h b/include/net/flow_dissector.h index 22aba321282d..9a074776f70b 100644 --- a/include/net/flow_dissector.h +++ b/include/net/flow_dissector.h @@ -84,11 +84,11 @@ struct flow_dissector_key_ipv6_addrs { }; /** - * struct flow_dissector_key_tipc_addrs: - * @srcnode: source node address + * struct flow_dissector_key_tipc: + * @key: source node address combined with selector */ -struct flow_dissector_key_tipc_addrs { - __be32 srcnode; +struct flow_dissector_key_tipc { + __be32 key; }; /** @@ -100,7 +100,7 @@ struct flow_dissector_key_addrs { union { struct flow_dissector_key_ipv4_addrs v4addrs; struct flow_dissector_key_ipv6_addrs v6addrs; - struct flow_dissector_key_tipc_addrs tipcaddrs; + struct flow_dissector_key_tipc tipckey; }; }; @@ -192,7 +192,7 @@ enum flow_dissector_key_id { FLOW_DISSECTOR_KEY_PORTS, /* struct flow_dissector_key_ports */ FLOW_DISSECTOR_KEY_ICMP, /* struct flow_dissector_key_icmp */ FLOW_DISSECTOR_KEY_ETH_ADDRS, /* struct flow_dissector_key_eth_addrs */ - FLOW_DISSECTOR_KEY_TIPC_ADDRS, /* struct flow_dissector_key_tipc_addrs */ + FLOW_DISSECTOR_KEY_TIPC, /* struct flow_dissector_key_tipc */ FLOW_DISSECTOR_KEY_ARP, /* struct flow_dissector_key_arp */ FLOW_DISSECTOR_KEY_VLAN, /* struct flow_dissector_key_flow_vlan */ FLOW_DISSECTOR_KEY_FLOW_LABEL, /* struct flow_dissector_key_flow_tags */ diff --git a/include/net/tipc.h b/include/net/tipc.h new file mode 100644 index 000000000000..07670ec022a7 --- /dev/null +++ b/include/net/tipc.h @@ -0,0 +1,62 @@ +/* + * include/net/tipc.h: Include file for TIPC message header routines + * + * Copyright (c) 2017 Ericsson AB + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_HDR_H +#define _TIPC_HDR_H + +#include + +#define KEEPALIVE_MSG_MASK 0x0e080000 /* LINK_PROTOCOL + MSG_IS_KEEPALIVE */ + +struct tipc_basic_hdr { + __be32 w[4]; +}; + +static inline u32 tipc_hdr_rps_key(struct tipc_basic_hdr *hdr) +{ + u32 w0 = ntohl(hdr->w[0]); + bool keepalive_msg = (w0 & KEEPALIVE_MSG_MASK) == KEEPALIVE_MSG_MASK; + int key; + + /* Return source node identity as key */ + if (likely(!keepalive_msg)) + return hdr->w[3]; + + /* Spread PROBE/PROBE_REPLY messages across the cores */ + get_random_bytes(&key, sizeof(key)); + return key; +} + +#endif diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c index 1f5caafb4492..15ce30063765 100644 --- a/net/core/flow_dissector.c +++ b/net/core/flow_dissector.c @@ -10,6 +10,7 @@ #include #include #include +#include #include #include #include @@ -772,23 +773,22 @@ proto_again: break; } case htons(ETH_P_TIPC): { - struct { - __be32 pre[3]; - __be32 srcnode; - } *hdr, _hdr; - hdr = __skb_header_pointer(skb, nhoff, sizeof(_hdr), data, hlen, &_hdr); + struct tipc_basic_hdr *hdr, _hdr; + + hdr = __skb_header_pointer(skb, nhoff, sizeof(_hdr), + data, hlen, &_hdr); if (!hdr) { fdret = FLOW_DISSECT_RET_OUT_BAD; break; } if (dissector_uses_key(flow_dissector, - FLOW_DISSECTOR_KEY_TIPC_ADDRS)) { + FLOW_DISSECTOR_KEY_TIPC)) { key_addrs = skb_flow_dissector_target(flow_dissector, - FLOW_DISSECTOR_KEY_TIPC_ADDRS, + FLOW_DISSECTOR_KEY_TIPC, target_container); - key_addrs->tipcaddrs.srcnode = hdr->srcnode; - key_control->addr_type = FLOW_DISSECTOR_KEY_TIPC_ADDRS; + key_addrs->tipckey.key = tipc_hdr_rps_key(hdr); + key_control->addr_type = FLOW_DISSECTOR_KEY_TIPC; } fdret = FLOW_DISSECT_RET_OUT_GOOD; break; @@ -1024,8 +1024,8 @@ static inline size_t flow_keys_hash_length(const struct flow_keys *flow) case FLOW_DISSECTOR_KEY_IPV6_ADDRS: diff -= sizeof(flow->addrs.v6addrs); break; - case FLOW_DISSECTOR_KEY_TIPC_ADDRS: - diff -= sizeof(flow->addrs.tipcaddrs); + case FLOW_DISSECTOR_KEY_TIPC: + diff -= sizeof(flow->addrs.tipckey); break; } return (sizeof(*flow) - diff) / sizeof(u32); @@ -1039,8 +1039,8 @@ __be32 flow_get_u32_src(const struct flow_keys *flow) case FLOW_DISSECTOR_KEY_IPV6_ADDRS: return (__force __be32)ipv6_addr_hash( &flow->addrs.v6addrs.src); - case FLOW_DISSECTOR_KEY_TIPC_ADDRS: - return flow->addrs.tipcaddrs.srcnode; + case FLOW_DISSECTOR_KEY_TIPC: + return flow->addrs.tipckey.key; default: return 0; } @@ -1321,8 +1321,8 @@ static const struct flow_dissector_key flow_keys_dissector_keys[] = { .offset = offsetof(struct flow_keys, addrs.v6addrs), }, { - .key_id = FLOW_DISSECTOR_KEY_TIPC_ADDRS, - .offset = offsetof(struct flow_keys, addrs.tipcaddrs), + .key_id = FLOW_DISSECTOR_KEY_TIPC, + .offset = offsetof(struct flow_keys, addrs.tipckey), }, { .key_id = FLOW_DISSECTOR_KEY_PORTS, diff --git a/net/tipc/link.c b/net/tipc/link.c index 870b9b8f877a..6bce0b1117bd 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -239,7 +239,8 @@ static int link_is_up(struct tipc_link *l) static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb, struct sk_buff_head *xmitq); static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe, - u16 rcvgap, int tolerance, int priority, + bool probe_reply, u16 rcvgap, + int tolerance, int priority, struct sk_buff_head *xmitq); static void link_print(struct tipc_link *l, const char *str); static int tipc_link_build_nack_msg(struct tipc_link *l, @@ -773,7 +774,7 @@ int tipc_link_timeout(struct tipc_link *l, struct sk_buff_head *xmitq) } if (state || probe || setup) - tipc_link_build_proto_msg(l, mtyp, probe, 0, 0, 0, xmitq); + tipc_link_build_proto_msg(l, mtyp, probe, 0, 0, 0, 0, xmitq); return rc; } @@ -1174,7 +1175,7 @@ int tipc_link_build_state_msg(struct tipc_link *l, struct sk_buff_head *xmitq) /* Unicast ACK */ l->rcv_unacked = 0; l->stats.sent_acks++; - tipc_link_build_proto_msg(l, STATE_MSG, 0, 0, 0, 0, xmitq); + tipc_link_build_proto_msg(l, STATE_MSG, 0, 0, 0, 0, 0, xmitq); return 0; } @@ -1188,7 +1189,7 @@ void tipc_link_build_reset_msg(struct tipc_link *l, struct sk_buff_head *xmitq) if (l->state == LINK_ESTABLISHING) mtyp = ACTIVATE_MSG; - tipc_link_build_proto_msg(l, mtyp, 0, 0, 0, 0, xmitq); + tipc_link_build_proto_msg(l, mtyp, 0, 0, 0, 0, 0, xmitq); /* Inform peer that this endpoint is going down if applicable */ skb = skb_peek_tail(xmitq); @@ -1215,7 +1216,7 @@ static int tipc_link_build_nack_msg(struct tipc_link *l, } if ((skb_queue_len(&l->deferdq) == 1) || !(def_cnt % TIPC_NACK_INTV)) - tipc_link_build_proto_msg(l, STATE_MSG, 0, 0, 0, 0, xmitq); + tipc_link_build_proto_msg(l, STATE_MSG, 0, 0, 0, 0, 0, xmitq); return 0; } @@ -1289,7 +1290,8 @@ drop: } static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe, - u16 rcvgap, int tolerance, int priority, + bool probe_reply, u16 rcvgap, + int tolerance, int priority, struct sk_buff_head *xmitq) { struct tipc_link *bcl = l->bc_rcvlink; @@ -1337,6 +1339,7 @@ static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe, msg_set_seq_gap(hdr, rcvgap); msg_set_bc_gap(hdr, link_bc_rcv_gap(bcl)); msg_set_probe(hdr, probe); + msg_set_is_keepalive(hdr, probe || probe_reply); tipc_mon_prep(l->net, data, &dlen, mstate, l->bearer_id); msg_set_size(hdr, INT_H_SIZE + dlen); skb_trim(skb, INT_H_SIZE + dlen); @@ -1442,6 +1445,7 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb, u16 rcv_nxt = l->rcv_nxt; u16 dlen = msg_data_sz(hdr); int mtyp = msg_type(hdr); + bool reply = msg_probe(hdr); void *data; char *if_name; int rc = 0; @@ -1528,9 +1532,9 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb, /* Send NACK if peer has sent pkts we haven't received yet */ if (more(peers_snd_nxt, rcv_nxt) && !tipc_link_is_synching(l)) rcvgap = peers_snd_nxt - l->rcv_nxt; - if (rcvgap || (msg_probe(hdr))) - tipc_link_build_proto_msg(l, STATE_MSG, 0, rcvgap, - 0, 0, xmitq); + if (rcvgap || reply) + tipc_link_build_proto_msg(l, STATE_MSG, 0, reply, + rcvgap, 0, 0, xmitq); tipc_link_release_pkts(l, ack); /* If NACK, retransmit will now start at right position */ @@ -2122,14 +2126,14 @@ void tipc_link_set_tolerance(struct tipc_link *l, u32 tol, struct sk_buff_head *xmitq) { l->tolerance = tol; - tipc_link_build_proto_msg(l, STATE_MSG, 0, 0, tol, 0, xmitq); + tipc_link_build_proto_msg(l, STATE_MSG, 0, 0, 0, tol, 0, xmitq); } void tipc_link_set_prio(struct tipc_link *l, u32 prio, struct sk_buff_head *xmitq) { l->priority = prio; - tipc_link_build_proto_msg(l, STATE_MSG, 0, 0, 0, prio, xmitq); + tipc_link_build_proto_msg(l, STATE_MSG, 0, 0, 0, 0, prio, xmitq); } void tipc_link_set_abort_limit(struct tipc_link *l, u32 limit) diff --git a/net/tipc/msg.h b/net/tipc/msg.h index cedf811317fb..bf8f57ccc70c 100644 --- a/net/tipc/msg.h +++ b/net/tipc/msg.h @@ -226,6 +226,16 @@ static inline void msg_set_dest_droppable(struct tipc_msg *m, u32 d) msg_set_bits(m, 0, 19, 1, d); } +static inline int msg_is_keepalive(struct tipc_msg *m) +{ + return msg_bits(m, 0, 19, 1); +} + +static inline void msg_set_is_keepalive(struct tipc_msg *m, u32 d) +{ + msg_set_bits(m, 0, 19, 1, d); +} + static inline int msg_src_droppable(struct tipc_msg *m) { return msg_bits(m, 0, 18, 1); -- cgit v1.2.3 From 713bafea92920103cd3d361657406cf04d0e22dd Mon Sep 17 00:00:00 2001 From: Yuchung Cheng Date: Wed, 8 Nov 2017 13:01:26 -0800 Subject: tcp: retire FACK loss detection FACK loss detection has been disabled by default and the successor RACK subsumed FACK and can handle reordering better. This patch removes FACK to simplify TCP loss recovery. Signed-off-by: Yuchung Cheng Reviewed-by: Eric Dumazet Reviewed-by: Neal Cardwell Reviewed-by: Soheil Hassas Yeganeh Reviewed-by: Priyaranjan Jha Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 3 +- include/linux/tcp.h | 1 - include/net/tcp.h | 14 +-------- include/uapi/linux/snmp.h | 1 - net/ipv4/proc.c | 1 - net/ipv4/tcp.c | 2 -- net/ipv4/tcp_input.c | 53 +++++----------------------------- net/ipv4/tcp_metrics.c | 4 +-- net/ipv4/tcp_minisocks.c | 5 +--- net/ipv4/tcp_output.c | 5 +--- 10 files changed, 12 insertions(+), 77 deletions(-) (limited to 'include/net') diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index d8676dda7fa6..46c7e1085efc 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -289,8 +289,7 @@ tcp_ecn_fallback - BOOLEAN Default: 1 (fallback enabled) tcp_fack - BOOLEAN - Enable FACK congestion avoidance and fast retransmission. - The value is not used, if tcp_sack is not enabled. + This is a legacy option, it has no effect anymore. tcp_fin_timeout - INTEGER The length of time an orphaned (no longer referenced by any diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 22f40c96a15b..9574936fe041 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -85,7 +85,6 @@ struct tcp_sack_block { /*These are used to set the sack_ok field in struct tcp_options_received */ #define TCP_SACK_SEEN (1 << 0) /*1 = peer is SACK capable, */ -#define TCP_FACK_ENABLED (1 << 1) /*1 = FACK is enabled locally*/ #define TCP_DSACK_SEEN (1 << 2) /*1 = DSACK was received from peer*/ struct tcp_options_received { diff --git a/include/net/tcp.h b/include/net/tcp.h index 2f2c69ad31b2..ed71511e67a6 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -384,7 +384,6 @@ void tcp_update_metrics(struct sock *sk); void tcp_init_metrics(struct sock *sk); void tcp_metrics_init(void); bool tcp_peer_is_proven(struct request_sock *req, struct dst_entry *dst); -void tcp_disable_fack(struct tcp_sock *tp); void tcp_close(struct sock *sk, long timeout); void tcp_init_sock(struct sock *sk); void tcp_init_transfer(struct sock *sk, int bpf_op); @@ -776,7 +775,7 @@ struct tcp_skb_cb { }; __u8 tcp_flags; /* TCP header flags. (tcp[13]) */ - __u8 sacked; /* State flags for SACK/FACK. */ + __u8 sacked; /* State flags for SACK. */ #define TCPCB_SACKED_ACKED 0x01 /* SKB ACK'd by a SACK block */ #define TCPCB_SACKED_RETRANS 0x02 /* SKB retransmitted */ #define TCPCB_LOST 0x04 /* SKB is lost */ @@ -1066,7 +1065,6 @@ void tcp_rate_check_app_limited(struct sock *sk); * * tcp_is_sack - SACK enabled * tcp_is_reno - No SACK - * tcp_is_fack - FACK enabled, implies SACK enabled */ static inline int tcp_is_sack(const struct tcp_sock *tp) { @@ -1078,16 +1076,6 @@ static inline bool tcp_is_reno(const struct tcp_sock *tp) return !tcp_is_sack(tp); } -static inline bool tcp_is_fack(const struct tcp_sock *tp) -{ - return tp->rx_opt.sack_ok & TCP_FACK_ENABLED; -} - -static inline void tcp_enable_fack(struct tcp_sock *tp) -{ - tp->rx_opt.sack_ok |= TCP_FACK_ENABLED; -} - static inline unsigned int tcp_left_out(const struct tcp_sock *tp) { return tp->sacked_out + tp->lost_out; diff --git a/include/uapi/linux/snmp.h b/include/uapi/linux/snmp.h index 0d941cdd8e8c..33a70ece462f 100644 --- a/include/uapi/linux/snmp.h +++ b/include/uapi/linux/snmp.h @@ -191,7 +191,6 @@ enum LINUX_MIB_TCPRENORECOVERY, /* TCPRenoRecovery */ LINUX_MIB_TCPSACKRECOVERY, /* TCPSackRecovery */ LINUX_MIB_TCPSACKRENEGING, /* TCPSACKReneging */ - LINUX_MIB_TCPFACKREORDER, /* TCPFACKReorder */ LINUX_MIB_TCPSACKREORDER, /* TCPSACKReorder */ LINUX_MIB_TCPRENOREORDER, /* TCPRenoReorder */ LINUX_MIB_TCPTSREORDER, /* TCPTSReorder */ diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c index 127153f1ed8a..9f37c4727861 100644 --- a/net/ipv4/proc.c +++ b/net/ipv4/proc.c @@ -212,7 +212,6 @@ static const struct snmp_mib snmp4_net_list[] = { SNMP_MIB_ITEM("TCPRenoRecovery", LINUX_MIB_TCPRENORECOVERY), SNMP_MIB_ITEM("TCPSackRecovery", LINUX_MIB_TCPSACKRECOVERY), SNMP_MIB_ITEM("TCPSACKReneging", LINUX_MIB_TCPSACKRENEGING), - SNMP_MIB_ITEM("TCPFACKReorder", LINUX_MIB_TCPFACKREORDER), SNMP_MIB_ITEM("TCPSACKReorder", LINUX_MIB_TCPSACKREORDER), SNMP_MIB_ITEM("TCPRenoReorder", LINUX_MIB_TCPRENOREORDER), SNMP_MIB_ITEM("TCPTSReorder", LINUX_MIB_TCPTSREORDER), diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index bc71a27d5ad9..337555076043 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2509,8 +2509,6 @@ static int tcp_repair_options_est(struct sock *sk, return -EINVAL; tp->rx_opt.sack_ok |= TCP_SACK_SEEN; - if (sock_net(sk)->ipv4.sysctl_tcp_fack) - tcp_enable_fack(tp); break; case TCPOPT_TIMESTAMP: if (opt.opt_val != 0) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 9ceaa1fdc3ab..487e181cff86 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -842,18 +842,6 @@ __u32 tcp_init_cwnd(const struct tcp_sock *tp, const struct dst_entry *dst) return min_t(__u32, cwnd, tp->snd_cwnd_clamp); } -/* - * Packet counting of FACK is based on in-order assumptions, therefore TCP - * disables it when reordering is detected - */ -void tcp_disable_fack(struct tcp_sock *tp) -{ - /* RFC3517 uses different metric in lost marker => reset on change */ - if (tcp_is_fack(tp)) - tp->lost_skb_hint = NULL; - tp->rx_opt.sack_ok &= ~TCP_FACK_ENABLED; -} - /* Take a notice that peer is sending D-SACKs */ static void tcp_dsack_seen(struct tcp_sock *tp) { @@ -881,7 +869,6 @@ static void tcp_update_reordering(struct sock *sk, const int metric, tp->sacked_out, tp->undo_marker ? tp->undo_retrans : 0); #endif - tcp_disable_fack(tp); } tp->rack.reord = 1; @@ -891,8 +878,6 @@ static void tcp_update_reordering(struct sock *sk, const int metric, mib_idx = LINUX_MIB_TCPTSREORDER; else if (tcp_is_reno(tp)) mib_idx = LINUX_MIB_TCPRENOREORDER; - else if (tcp_is_fack(tp)) - mib_idx = LINUX_MIB_TCPFACKREORDER; else mib_idx = LINUX_MIB_TCPSACKREORDER; @@ -970,7 +955,6 @@ void tcp_skb_mark_lost_uncond_verify(struct tcp_sock *tp, struct sk_buff *skb) * 3. Loss detection event of two flavors: * A. Scoreboard estimator decided the packet is lost. * A'. Reno "three dupacks" marks head of queue lost. - * A''. Its FACK modification, head until snd.fack is lost. * B. SACK arrives sacking SND.NXT at the moment, when the * segment was retransmitted. * 4. D-SACK added new rule: D-SACK changes any tag to S. @@ -1248,7 +1232,7 @@ static u8 tcp_sacktag_one(struct sock *sk, fack_count += pcount; /* Lost marker hint past SACKed? Tweak RFC3517 cnt */ - if (!tcp_is_fack(tp) && tp->lost_skb_hint && + if (tp->lost_skb_hint && before(start_seq, TCP_SKB_CB(tp->lost_skb_hint)->seq)) tp->lost_cnt_hint += pcount; @@ -2051,10 +2035,6 @@ static inline int tcp_fackets_out(const struct tcp_sock *tp) * counter when SACK is enabled (without SACK, sacked_out is used for * that purpose). * - * Instead, with FACK TCP uses fackets_out that includes both SACKed - * segments up to the highest received SACK block so far and holes in - * between them. - * * With reordering, holes may still be in flight, so RFC3517 recovery * uses pure sacked_out (total number of SACKed segments) even though * it violates the RFC that uses duplicate ACKs, often these are equal @@ -2064,10 +2044,10 @@ static inline int tcp_fackets_out(const struct tcp_sock *tp) */ static inline int tcp_dupack_heuristics(const struct tcp_sock *tp) { - return tcp_is_fack(tp) ? tp->fackets_out : tp->sacked_out + 1; + return tp->sacked_out + 1; } -/* Linux NewReno/SACK/FACK/ECN state machine. +/* Linux NewReno/SACK/ECN state machine. * -------------------------------------- * * "Open" Normal state, no dubious events, fast path. @@ -2132,16 +2112,6 @@ static inline int tcp_dupack_heuristics(const struct tcp_sock *tp) * dynamically measured and adjusted. This is implemented in * tcp_rack_mark_lost. * - * FACK (Disabled by default. Subsumbed by RACK): - * It is the simplest heuristics. As soon as we decided - * that something is lost, we decide that _all_ not SACKed - * packets until the most forward SACK are lost. I.e. - * lost_out = fackets_out - sacked_out and left_out = fackets_out. - * It is absolutely correct estimate, if network does not reorder - * packets. And it loses any connection to reality when reordering - * takes place. We use FACK by default until reordering - * is suspected on the path to this destination. - * * If the receiver does not support SACK: * * NewReno (RFC6582): in Recovery we assume that one segment @@ -2190,7 +2160,7 @@ static bool tcp_time_to_recover(struct sock *sk, int flag) } /* Detect loss in event "A" above by marking head of queue up as lost. - * For FACK or non-SACK(Reno) senders, the first "packets" number of segments + * For non-SACK(Reno) senders, the first "packets" number of segments * are considered lost. For RFC3517 SACK, a segment is considered lost if it * has at least tp->reordering SACKed seqments above it; "packets" refers to * the maximum SACKed segments to pass before reaching this limit. @@ -2226,12 +2196,12 @@ static void tcp_mark_head_lost(struct sock *sk, int packets, int mark_head) break; oldcnt = cnt; - if (tcp_is_fack(tp) || tcp_is_reno(tp) || + if (tcp_is_reno(tp) || (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) cnt += tcp_skb_pcount(skb); if (cnt > packets) { - if ((tcp_is_sack(tp) && !tcp_is_fack(tp)) || + if (tcp_is_sack(tp) || (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED) || (oldcnt >= packets)) break; @@ -2262,11 +2232,6 @@ static void tcp_update_scoreboard(struct sock *sk, int fast_rexmit) if (tcp_is_reno(tp)) { tcp_mark_head_lost(sk, 1, 1); - } else if (tcp_is_fack(tp)) { - int lost = tp->fackets_out - tp->reordering; - if (lost <= 0) - lost = 1; - tcp_mark_head_lost(sk, lost, 0); } else { int sacked_upto = tp->sacked_out - tp->reordering; if (sacked_upto >= 0) @@ -3199,8 +3164,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, if (reord < prior_fackets && reord <= tp->fackets_out) tcp_update_reordering(sk, tp->fackets_out - reord, 0); - delta = tcp_is_fack(tp) ? pkts_acked : - prior_sacked - tp->sacked_out; + delta = prior_sacked - tp->sacked_out; tp->lost_cnt_hint -= min(tp->lost_cnt_hint, delta); } @@ -5708,9 +5672,6 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, tp->tcp_header_len = sizeof(struct tcphdr); } - if (tcp_is_sack(tp) && sock_net(sk)->ipv4.sysctl_tcp_fack) - tcp_enable_fack(tp); - tcp_sync_mss(sk, icsk->icsk_pmtu_cookie); tcp_initialize_rcv_mss(sk); diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp_metrics.c index 9d5ddebfd831..7097f92d16e5 100644 --- a/net/ipv4/tcp_metrics.c +++ b/net/ipv4/tcp_metrics.c @@ -470,10 +470,8 @@ void tcp_init_metrics(struct sock *sk) tp->snd_ssthresh = TCP_INFINITE_SSTHRESH; } val = tcp_metric_get(tm, TCP_METRIC_REORDERING); - if (val && tp->reordering != val) { - tcp_disable_fack(tp); + if (val && tp->reordering != val) tp->reordering = val; - } crtt = tcp_metric_get(tm, TCP_METRIC_RTT); rcu_read_unlock(); diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 4bb86580decd..326c9282bf94 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -509,10 +509,7 @@ struct sock *tcp_create_openreq_child(const struct sock *sk, keepalive_time_when(newtp)); newtp->rx_opt.tstamp_ok = ireq->tstamp_ok; - if ((newtp->rx_opt.sack_ok = ireq->sack_ok) != 0) { - if (sock_net(sk)->ipv4.sysctl_tcp_fack) - tcp_enable_fack(newtp); - } + newtp->rx_opt.sack_ok = ireq->sack_ok; newtp->window_clamp = req->rsk_window_clamp; newtp->rcv_ssthresh = req->rsk_rcv_wnd; newtp->rcv_wnd = req->rsk_rcv_wnd; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 9b98d35aa0d8..094c429b4401 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1257,7 +1257,7 @@ static void tcp_adjust_pcount(struct sock *sk, const struct sk_buff *skb, int de if (tp->lost_skb_hint && before(TCP_SKB_CB(skb)->seq, TCP_SKB_CB(tp->lost_skb_hint)->seq) && - (tcp_is_fack(tp) || (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED))) + (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) tp->lost_cnt_hint -= decr; tcp_verify_left_out(tp); @@ -2961,9 +2961,6 @@ int tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb, int segs) * retransmitted data is acknowledged. It tries to continue * resending the rest of the retransmit queue, until either * we've sent it all or the congestion window limit is reached. - * If doing SACK, the first ACK which comes back for a timeout - * based retransmit packet might feed us FACK information again. - * If so, we use it to avoid unnecessarily retransmissions. */ void tcp_xmit_retransmit_queue(struct sock *sk) { -- cgit v1.2.3 From 39b175211053c7a6a4d794c42e225994f1c069c2 Mon Sep 17 00:00:00 2001 From: Mat Martineau Date: Fri, 10 Nov 2017 14:03:51 -0800 Subject: net: Remove unused skb_shared_info member ip6_frag_id was only used by UFO, which has been removed. ipv6_proxy_select_ident() only existed to set ip6_frag_id and has no in-tree callers. Signed-off-by: Mat Martineau Signed-off-by: David S. Miller --- include/linux/skbuff.h | 1 - include/net/ipv6.h | 1 - net/ipv6/output_core.c | 31 ------------------------------- 3 files changed, 33 deletions(-) (limited to 'include/net') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 57d712671081..54fe91183a8e 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -500,7 +500,6 @@ struct skb_shared_info { struct skb_shared_hwtstamps hwtstamps; unsigned int gso_type; u32 tskey; - __be32 ip6_frag_id; /* * Warning : all fields before dataref are cleared in __alloc_skb() diff --git a/include/net/ipv6.h b/include/net/ipv6.h index fb6d67012de6..ec14f0d5a3a1 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -767,7 +767,6 @@ static inline int ipv6_addr_diff(const struct in6_addr *a1, const struct in6_add __be32 ipv6_select_ident(struct net *net, const struct in6_addr *daddr, const struct in6_addr *saddr); -void ipv6_proxy_select_ident(struct net *net, struct sk_buff *skb); int ip6_dst_hoplimit(struct dst_entry *dst); diff --git a/net/ipv6/output_core.c b/net/ipv6/output_core.c index a338bbc33cf3..4a7e5ffa5108 100644 --- a/net/ipv6/output_core.c +++ b/net/ipv6/output_core.c @@ -31,37 +31,6 @@ static u32 __ipv6_select_ident(struct net *net, u32 hashrnd, return id; } -/* This function exists only for tap drivers that must support broken - * clients requesting UFO without specifying an IPv6 fragment ID. - * - * This is similar to ipv6_select_ident() but we use an independent hash - * seed to limit information leakage. - * - * The network header must be set before calling this. - */ -void ipv6_proxy_select_ident(struct net *net, struct sk_buff *skb) -{ - static u32 ip6_proxy_idents_hashrnd __read_mostly; - struct in6_addr buf[2]; - struct in6_addr *addrs; - u32 id; - - addrs = skb_header_pointer(skb, - skb_network_offset(skb) + - offsetof(struct ipv6hdr, saddr), - sizeof(buf), buf); - if (!addrs) - return; - - net_get_random_once(&ip6_proxy_idents_hashrnd, - sizeof(ip6_proxy_idents_hashrnd)); - - id = __ipv6_select_ident(net, ip6_proxy_idents_hashrnd, - &addrs[1], &addrs[0]); - skb_shinfo(skb)->ip6_frag_id = htonl(id); -} -EXPORT_SYMBOL_GPL(ipv6_proxy_select_ident); - __be32 ipv6_select_ident(struct net *net, const struct in6_addr *daddr, const struct in6_addr *saddr) -- cgit v1.2.3 From 5ed4e3eb021762fee584ce65620bc822131c7aa0 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Fri, 10 Nov 2017 15:22:52 -0800 Subject: net: dsa: Pass a port to get_tag_protocol() A number of drivers want to check whether the configured CPU port is a possible configuration for enabling tagging, pass down the CPU port number so they verify that. Signed-off-by: Florian Fainelli Reviewed-by: Vivien Didelot Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- drivers/net/dsa/b53/b53_common.c | 32 +++++++++++++------------------- drivers/net/dsa/bcm_sf2.c | 3 ++- drivers/net/dsa/dsa_loop.c | 3 ++- drivers/net/dsa/lan9303-core.c | 3 ++- drivers/net/dsa/microchip/ksz_common.c | 3 ++- drivers/net/dsa/mt7530.c | 4 ++-- drivers/net/dsa/mv88e6060.c | 3 ++- drivers/net/dsa/mv88e6xxx/chip.c | 3 ++- drivers/net/dsa/qca8k.c | 2 +- include/net/dsa.h | 3 ++- net/dsa/dsa2.c | 2 +- net/dsa/legacy.c | 2 +- 12 files changed, 32 insertions(+), 31 deletions(-) (limited to 'include/net') diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c index 44a9a03bff55..f72aeb9ed12a 100644 --- a/drivers/net/dsa/b53/b53_common.c +++ b/drivers/net/dsa/b53/b53_common.c @@ -541,7 +541,8 @@ EXPORT_SYMBOL(b53_disable_port); void b53_brcm_hdr_setup(struct dsa_switch *ds, int port) { - bool tag_en = !!(ds->ops->get_tag_protocol(ds) == DSA_TAG_PROTO_BRCM); + bool tag_en = !!(ds->ops->get_tag_protocol(ds, port) == + DSA_TAG_PROTO_BRCM); struct b53_device *dev = ds->priv; u8 hdr_ctl, val; u16 reg; @@ -1478,38 +1479,31 @@ void b53_br_fast_age(struct dsa_switch *ds, int port) } EXPORT_SYMBOL(b53_br_fast_age); -static bool b53_can_enable_brcm_tags(struct dsa_switch *ds) +static bool b53_can_enable_brcm_tags(struct dsa_switch *ds, int port) { - unsigned int brcm_tag_mask; - unsigned int i; - /* Broadcom switches will accept enabling Broadcom tags on the * following ports: 5, 7 and 8, any other port is not supported */ - brcm_tag_mask = BIT(B53_CPU_PORT_25) | BIT(7) | BIT(B53_CPU_PORT); - - for (i = 0; i < ds->num_ports; i++) { - if (dsa_is_cpu_port(ds, i)) { - if (!(BIT(i) & brcm_tag_mask)) { - dev_warn(ds->dev, - "Port %d is not Broadcom tag capable\n", - i); - return false; - } - } + switch (port) { + case B53_CPU_PORT_25: + case 7: + case B53_CPU_PORT: + return true; } - return true; + dev_warn(ds->dev, "Port %d is not Broadcom tag capable\n", port); + return false; } -static enum dsa_tag_protocol b53_get_tag_protocol(struct dsa_switch *ds) +static enum dsa_tag_protocol b53_get_tag_protocol(struct dsa_switch *ds, + int port) { struct b53_device *dev = ds->priv; /* Older models support a different tag format that we do not * support in net/dsa/tag_brcm.c yet. */ - if (is5325(dev) || is5365(dev) || !b53_can_enable_brcm_tags(ds)) + if (is5325(dev) || is5365(dev) || !b53_can_enable_brcm_tags(ds, port)) return DSA_TAG_PROTO_NONE; else return DSA_TAG_PROTO_BRCM; diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c index 2d6867f4008c..93faa1fed6f2 100644 --- a/drivers/net/dsa/bcm_sf2.c +++ b/drivers/net/dsa/bcm_sf2.c @@ -35,7 +35,8 @@ #include "b53/b53_priv.h" #include "b53/b53_regs.h" -static enum dsa_tag_protocol bcm_sf2_sw_get_tag_protocol(struct dsa_switch *ds) +static enum dsa_tag_protocol bcm_sf2_sw_get_tag_protocol(struct dsa_switch *ds, + int port) { return DSA_TAG_PROTO_BRCM; } diff --git a/drivers/net/dsa/dsa_loop.c b/drivers/net/dsa/dsa_loop.c index 3a3f4f7ba364..bb71d3d6f65b 100644 --- a/drivers/net/dsa/dsa_loop.c +++ b/drivers/net/dsa/dsa_loop.c @@ -64,7 +64,8 @@ struct dsa_loop_priv { static struct phy_device *phydevs[PHY_MAX_ADDR]; -static enum dsa_tag_protocol dsa_loop_get_protocol(struct dsa_switch *ds) +static enum dsa_tag_protocol dsa_loop_get_protocol(struct dsa_switch *ds, + int port) { dev_dbg(ds->dev, "%s\n", __func__); diff --git a/drivers/net/dsa/lan9303-core.c b/drivers/net/dsa/lan9303-core.c index a2610085e7ba..fdfdb0edfe62 100644 --- a/drivers/net/dsa/lan9303-core.c +++ b/drivers/net/dsa/lan9303-core.c @@ -894,7 +894,8 @@ static int lan9303_check_device(struct lan9303 *chip) /* ---------------------------- DSA -----------------------------------*/ -static enum dsa_tag_protocol lan9303_get_tag_protocol(struct dsa_switch *ds) +static enum dsa_tag_protocol lan9303_get_tag_protocol(struct dsa_switch *ds, + int port) { return DSA_TAG_PROTO_LAN9303; } diff --git a/drivers/net/dsa/microchip/ksz_common.c b/drivers/net/dsa/microchip/ksz_common.c index 56cd6d365352..b5be93a1e0df 100644 --- a/drivers/net/dsa/microchip/ksz_common.c +++ b/drivers/net/dsa/microchip/ksz_common.c @@ -394,7 +394,8 @@ static int ksz_setup(struct dsa_switch *ds) return 0; } -static enum dsa_tag_protocol ksz_get_tag_protocol(struct dsa_switch *ds) +static enum dsa_tag_protocol ksz_get_tag_protocol(struct dsa_switch *ds, + int port) { return DSA_TAG_PROTO_KSZ; } diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c index 627c039f12ca..2820d69810b3 100644 --- a/drivers/net/dsa/mt7530.c +++ b/drivers/net/dsa/mt7530.c @@ -907,11 +907,11 @@ err: } static enum dsa_tag_protocol -mtk_get_tag_protocol(struct dsa_switch *ds) +mtk_get_tag_protocol(struct dsa_switch *ds, int port) { struct mt7530_priv *priv = ds->priv; - if (!dsa_is_cpu_port(ds, MT7530_CPU_PORT)) { + if (port != MT7530_CPU_PORT) { dev_warn(priv->dev, "port not matched with tagging CPU port\n"); return DSA_TAG_PROTO_NONE; diff --git a/drivers/net/dsa/mv88e6060.c b/drivers/net/dsa/mv88e6060.c index 45768e3c5bc5..65f10fec25b3 100644 --- a/drivers/net/dsa/mv88e6060.c +++ b/drivers/net/dsa/mv88e6060.c @@ -70,7 +70,8 @@ static const char *mv88e6060_get_name(struct mii_bus *bus, int sw_addr) return NULL; } -static enum dsa_tag_protocol mv88e6060_get_tag_protocol(struct dsa_switch *ds) +static enum dsa_tag_protocol mv88e6060_get_tag_protocol(struct dsa_switch *ds, + int port) { return DSA_TAG_PROTO_TRAILER; } diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c index d6c3a22c8789..8171055fde7a 100644 --- a/drivers/net/dsa/mv88e6xxx/chip.c +++ b/drivers/net/dsa/mv88e6xxx/chip.c @@ -3731,7 +3731,8 @@ static int mv88e6xxx_smi_init(struct mv88e6xxx_chip *chip, return 0; } -static enum dsa_tag_protocol mv88e6xxx_get_tag_protocol(struct dsa_switch *ds) +static enum dsa_tag_protocol mv88e6xxx_get_tag_protocol(struct dsa_switch *ds, + int port) { struct mv88e6xxx_chip *chip = ds->priv; diff --git a/drivers/net/dsa/qca8k.c b/drivers/net/dsa/qca8k.c index cf72e274275f..9df22ebee822 100644 --- a/drivers/net/dsa/qca8k.c +++ b/drivers/net/dsa/qca8k.c @@ -823,7 +823,7 @@ qca8k_port_fdb_dump(struct dsa_switch *ds, int port, } static enum dsa_tag_protocol -qca8k_get_tag_protocol(struct dsa_switch *ds) +qca8k_get_tag_protocol(struct dsa_switch *ds, int port) { return DSA_TAG_PROTO_QCA; } diff --git a/include/net/dsa.h b/include/net/dsa.h index 6c239257309b..68e232fd4b0f 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -321,7 +321,8 @@ struct dsa_switch_ops { struct device *host_dev, int sw_addr, void **priv); - enum dsa_tag_protocol (*get_tag_protocol)(struct dsa_switch *ds); + enum dsa_tag_protocol (*get_tag_protocol)(struct dsa_switch *ds, + int port); int (*setup)(struct dsa_switch *ds); u32 (*get_phy_flags)(struct dsa_switch *ds, int port); diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c index fd54a8e17986..44e3fb7dec8c 100644 --- a/net/dsa/dsa2.c +++ b/net/dsa/dsa2.c @@ -539,7 +539,7 @@ static int dsa_port_parse_cpu(struct dsa_port *dp, struct net_device *master) const struct dsa_device_ops *tag_ops; enum dsa_tag_protocol tag_protocol; - tag_protocol = ds->ops->get_tag_protocol(ds); + tag_protocol = ds->ops->get_tag_protocol(ds, dp->index); tag_ops = dsa_resolve_tag_protocol(tag_protocol); if (IS_ERR(tag_ops)) { dev_warn(ds->dev, "No tagger for this switch\n"); diff --git a/net/dsa/legacy.c b/net/dsa/legacy.c index 4863e3e398b6..84611d7fcfa2 100644 --- a/net/dsa/legacy.c +++ b/net/dsa/legacy.c @@ -151,7 +151,7 @@ static int dsa_switch_setup_one(struct dsa_switch *ds, const struct dsa_device_ops *tag_ops; enum dsa_tag_protocol tag_protocol; - tag_protocol = ops->get_tag_protocol(ds); + tag_protocol = ops->get_tag_protocol(ds, dst->cpu_dp->index); tag_ops = dsa_resolve_tag_protocol(tag_protocol); if (IS_ERR(tag_ops)) return PTR_ERR(tag_ops); -- cgit v1.2.3 From b74b70c44986dee87881fbed3d912e02c5dcf78c Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Fri, 10 Nov 2017 15:22:54 -0800 Subject: net: dsa: Support prepended Broadcom tag Add a new type: DSA_TAG_PROTO_PREPEND which allows us to support for the 4-bytes Broadcom tag that we already support, but in a format where it is pre-pended to the packet instead of located between the MAC SA and the Ethertyper (DSA_TAG_PROTO_BRCM). Signed-off-by: Florian Fainelli Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- include/net/dsa.h | 1 + net/dsa/Kconfig | 3 +++ net/dsa/Makefile | 1 + net/dsa/dsa.c | 3 +++ net/dsa/dsa_priv.h | 1 + net/dsa/tag_brcm.c | 39 ++++++++++++++++++++++++++++++++------- 6 files changed, 41 insertions(+), 7 deletions(-) (limited to 'include/net') diff --git a/include/net/dsa.h b/include/net/dsa.h index 68e232fd4b0f..2a05738570d8 100644 --- a/include/net/dsa.h +++ b/include/net/dsa.h @@ -29,6 +29,7 @@ struct fixed_phy_status; enum dsa_tag_protocol { DSA_TAG_PROTO_NONE = 0, DSA_TAG_PROTO_BRCM, + DSA_TAG_PROTO_BRCM_PREPEND, DSA_TAG_PROTO_DSA, DSA_TAG_PROTO_EDSA, DSA_TAG_PROTO_KSZ, diff --git a/net/dsa/Kconfig b/net/dsa/Kconfig index cc5f8f971689..2fed892094bc 100644 --- a/net/dsa/Kconfig +++ b/net/dsa/Kconfig @@ -19,6 +19,9 @@ if NET_DSA config NET_DSA_TAG_BRCM bool +config NET_DSA_TAG_BRCM_PREPEND + bool + config NET_DSA_TAG_DSA bool diff --git a/net/dsa/Makefile b/net/dsa/Makefile index e9a4a0f33e86..0e13c1f95d13 100644 --- a/net/dsa/Makefile +++ b/net/dsa/Makefile @@ -5,6 +5,7 @@ dsa_core-y += dsa.o dsa2.o legacy.o master.o port.o slave.o switch.o # tagging formats dsa_core-$(CONFIG_NET_DSA_TAG_BRCM) += tag_brcm.o +dsa_core-$(CONFIG_NET_DSA_TAG_BRCM_PREPEND) += tag_brcm.o dsa_core-$(CONFIG_NET_DSA_TAG_DSA) += tag_dsa.o dsa_core-$(CONFIG_NET_DSA_TAG_EDSA) += tag_edsa.o dsa_core-$(CONFIG_NET_DSA_TAG_KSZ) += tag_ksz.o diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c index b8f2d9f7c3ed..6a9d0f50fbee 100644 --- a/net/dsa/dsa.c +++ b/net/dsa/dsa.c @@ -44,6 +44,9 @@ const struct dsa_device_ops *dsa_device_ops[DSA_TAG_LAST] = { #ifdef CONFIG_NET_DSA_TAG_BRCM [DSA_TAG_PROTO_BRCM] = &brcm_netdev_ops, #endif +#ifdef CONFIG_NET_DSA_TAG_BRCM_PREPEND + [DSA_TAG_PROTO_BRCM_PREPEND] = &brcm_prepend_netdev_ops, +#endif #ifdef CONFIG_NET_DSA_TAG_DSA [DSA_TAG_PROTO_DSA] = &dsa_netdev_ops, #endif diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h index 507e1ce4d4d2..7d036696e8c4 100644 --- a/net/dsa/dsa_priv.h +++ b/net/dsa/dsa_priv.h @@ -191,6 +191,7 @@ void dsa_switch_unregister_notifier(struct dsa_switch *ds); /* tag_brcm.c */ extern const struct dsa_device_ops brcm_netdev_ops; +extern const struct dsa_device_ops brcm_prepend_netdev_ops; /* tag_dsa.c */ extern const struct dsa_device_ops dsa_netdev_ops; diff --git a/net/dsa/tag_brcm.c b/net/dsa/tag_brcm.c index 771409a1e65c..e6e0b7b6025c 100644 --- a/net/dsa/tag_brcm.c +++ b/net/dsa/tag_brcm.c @@ -96,13 +96,6 @@ static struct sk_buff *brcm_tag_xmit_ll(struct sk_buff *skb, return skb; } -static struct sk_buff *brcm_tag_xmit(struct sk_buff *skb, - struct net_device *dev) -{ - /* Build the tag after the MAC Source Address */ - return brcm_tag_xmit_ll(skb, dev, 2 * ETH_ALEN); -} - static struct sk_buff *brcm_tag_rcv_ll(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt, @@ -139,6 +132,15 @@ static struct sk_buff *brcm_tag_rcv_ll(struct sk_buff *skb, return skb; } +#ifdef CONFIG_NET_DSA_TAG_BRCM +static struct sk_buff *brcm_tag_xmit(struct sk_buff *skb, + struct net_device *dev) +{ + /* Build the tag after the MAC Source Address */ + return brcm_tag_xmit_ll(skb, dev, 2 * ETH_ALEN); +} + + static struct sk_buff *brcm_tag_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt) { @@ -161,3 +163,26 @@ const struct dsa_device_ops brcm_netdev_ops = { .xmit = brcm_tag_xmit, .rcv = brcm_tag_rcv, }; +#endif + +#ifdef CONFIG_NET_DSA_TAG_BRCM_PREPEND +static struct sk_buff *brcm_tag_xmit_prepend(struct sk_buff *skb, + struct net_device *dev) +{ + /* tag is prepended to the packet */ + return brcm_tag_xmit_ll(skb, dev, 0); +} + +static struct sk_buff *brcm_tag_rcv_prepend(struct sk_buff *skb, + struct net_device *dev, + struct packet_type *pt) +{ + /* tag is prepended to the packet */ + return brcm_tag_rcv_ll(skb, dev, pt, ETH_HLEN); +} + +const struct dsa_device_ops brcm_prepend_netdev_ops = { + .xmit = brcm_tag_xmit_prepend, + .rcv = brcm_tag_rcv_prepend, +}; +#endif -- cgit v1.2.3 From 3a9b76fd0db9f0d426533f96a68a62a58753a51e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 11 Nov 2017 15:54:12 -0800 Subject: tcp: allow drivers to tweak TSQ logic MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit I had many reports that TSQ logic breaks wifi aggregation. Current logic is to allow up to 1 ms of bytes to be queued into qdisc and drivers queues. But Wifi aggregation needs a bigger budget to allow bigger rates to be discovered by various TCP Congestion Controls algorithms. This patch adds an extra socket field, allowing wifi drivers to select another log scale to derive TCP Small Queue credit from current pacing rate. Initial value is 10, meaning that this patch does not change current behavior. We expect wifi drivers to set this field to smaller values (tests have been done with values from 6 to 9) They would have to use following template : if (skb->sk && skb->sk->sk_pacing_shift != MY_PACING_SHIFT) skb->sk->sk_pacing_shift = MY_PACING_SHIFT; Ref: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1670041 Signed-off-by: Eric Dumazet Cc: Johannes Berg Cc: Toke Høiland-Jørgensen Cc: Kir Kolyshkin Acked-by: Neal Cardwell Signed-off-by: David S. Miller --- include/net/sock.h | 2 ++ net/core/sock.c | 1 + net/ipv4/tcp_output.c | 4 ++-- 3 files changed, 5 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index 688a823dccc3..f8715c5af37d 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -267,6 +267,7 @@ struct sock_common { * @sk_gso_type: GSO type (e.g. %SKB_GSO_TCPV4) * @sk_gso_max_size: Maximum GSO segment size to build * @sk_gso_max_segs: Maximum number of GSO segments + * @sk_pacing_shift: scaling factor for TCP Small Queues * @sk_lingertime: %SO_LINGER l_linger setting * @sk_backlog: always used with the per-socket spinlock held * @sk_callback_lock: used with the callbacks in the end of this struct @@ -451,6 +452,7 @@ struct sock { kmemcheck_bitfield_end(flags); u16 sk_gso_max_segs; + u8 sk_pacing_shift; unsigned long sk_lingertime; struct proto *sk_prot_creator; rwlock_t sk_callback_lock; diff --git a/net/core/sock.c b/net/core/sock.c index 57bbd6040eb6..13719af7b4e3 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -2746,6 +2746,7 @@ void sock_init_data(struct socket *sock, struct sock *sk) sk->sk_max_pacing_rate = ~0U; sk->sk_pacing_rate = ~0U; + sk->sk_pacing_shift = 10; sk->sk_incoming_cpu = -1; /* * Before updating sk_refcnt, we must commit prior changes to memory diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 0256f7a41041..76dbe884f246 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1720,7 +1720,7 @@ u32 tcp_tso_autosize(const struct sock *sk, unsigned int mss_now, { u32 bytes, segs; - bytes = min(sk->sk_pacing_rate >> 10, + bytes = min(sk->sk_pacing_rate >> sk->sk_pacing_shift, sk->sk_gso_max_size - 1 - MAX_TCP_HEADER); /* Goal is to send at least one packet per ms, @@ -2198,7 +2198,7 @@ static bool tcp_small_queue_check(struct sock *sk, const struct sk_buff *skb, { unsigned int limit; - limit = max(2 * skb->truesize, sk->sk_pacing_rate >> 10); + limit = max(2 * skb->truesize, sk->sk_pacing_rate >> sk->sk_pacing_shift); limit = min_t(u32, limit, sock_net(sk)->ipv4.sysctl_tcp_limit_output_bytes); limit <<= factor; -- cgit v1.2.3 From 6d88207fcfddc002afe3e2e4a455e5201089d5d9 Mon Sep 17 00:00:00 2001 From: Ilya Lesokhin Date: Mon, 13 Nov 2017 10:22:45 +0200 Subject: tls: Add function to update the TLS socket configuration The tx configuration is now stored in ctx->tx_conf. And sk->sk_prot is updated trough a function This will simplify things when we add rx and support for different possible tx and rx cross configurations. Signed-off-by: Ilya Lesokhin Signed-off-by: David S. Miller --- include/net/tls.h | 2 ++ net/tls/tls_main.c | 46 ++++++++++++++++++++++++++++++++-------------- 2 files changed, 34 insertions(+), 14 deletions(-) (limited to 'include/net') diff --git a/include/net/tls.h b/include/net/tls.h index b89d397dd62f..f058a6e08eaa 100644 --- a/include/net/tls.h +++ b/include/net/tls.h @@ -83,6 +83,8 @@ struct tls_context { void *priv_ctx; + u8 tx_conf:2; + u16 prepend_size; u16 tag_size; u16 overhead_size; diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c index 60aff60e30ad..de6a1416bc41 100644 --- a/net/tls/tls_main.c +++ b/net/tls/tls_main.c @@ -45,8 +45,18 @@ MODULE_AUTHOR("Mellanox Technologies"); MODULE_DESCRIPTION("Transport Layer Security Support"); MODULE_LICENSE("Dual BSD/GPL"); -static struct proto tls_base_prot; -static struct proto tls_sw_prot; +enum { + TLS_BASE_TX, + TLS_SW_TX, + TLS_NUM_CONFIG, +}; + +static struct proto tls_prots[TLS_NUM_CONFIG]; + +static inline void update_sk_prot(struct sock *sk, struct tls_context *ctx) +{ + sk->sk_prot = &tls_prots[ctx->tx_conf]; +} int wait_on_pending_writer(struct sock *sk, long *timeo) { @@ -340,8 +350,8 @@ static int do_tls_setsockopt_tx(struct sock *sk, char __user *optval, { struct tls_crypto_info *crypto_info, tmp_crypto_info; struct tls_context *ctx = tls_get_ctx(sk); - struct proto *prot = NULL; int rc = 0; + int tx_conf; if (!optval || (optlen < sizeof(*crypto_info))) { rc = -EINVAL; @@ -396,11 +406,12 @@ static int do_tls_setsockopt_tx(struct sock *sk, char __user *optval, /* currently SW is default, we will have ethtool in future */ rc = tls_set_sw_offload(sk, ctx); - prot = &tls_sw_prot; + tx_conf = TLS_SW_TX; if (rc) goto err_crypto_info; - sk->sk_prot = prot; + ctx->tx_conf = tx_conf; + update_sk_prot(sk, ctx); goto out; err_crypto_info: @@ -453,7 +464,9 @@ static int tls_init(struct sock *sk) icsk->icsk_ulp_data = ctx; ctx->setsockopt = sk->sk_prot->setsockopt; ctx->getsockopt = sk->sk_prot->getsockopt; - sk->sk_prot = &tls_base_prot; + + ctx->tx_conf = TLS_BASE_TX; + update_sk_prot(sk, ctx); out: return rc; } @@ -464,16 +477,21 @@ static struct tcp_ulp_ops tcp_tls_ulp_ops __read_mostly = { .init = tls_init, }; +static void build_protos(struct proto *prot, struct proto *base) +{ + prot[TLS_BASE_TX] = *base; + prot[TLS_BASE_TX].setsockopt = tls_setsockopt; + prot[TLS_BASE_TX].getsockopt = tls_getsockopt; + + prot[TLS_SW_TX] = prot[TLS_BASE_TX]; + prot[TLS_SW_TX].close = tls_sk_proto_close; + prot[TLS_SW_TX].sendmsg = tls_sw_sendmsg; + prot[TLS_SW_TX].sendpage = tls_sw_sendpage; +} + static int __init tls_register(void) { - tls_base_prot = tcp_prot; - tls_base_prot.setsockopt = tls_setsockopt; - tls_base_prot.getsockopt = tls_getsockopt; - - tls_sw_prot = tls_base_prot; - tls_sw_prot.sendmsg = tls_sw_sendmsg; - tls_sw_prot.sendpage = tls_sw_sendpage; - tls_sw_prot.close = tls_sk_proto_close; + build_protos(tls_prots, &tcp_prot); tcp_register_ulp(&tcp_tls_ulp_ops); -- cgit v1.2.3 From ff45d820a2df163957ad8ab459b6eb6976144c18 Mon Sep 17 00:00:00 2001 From: Ilya Lesokhin Date: Mon, 13 Nov 2017 10:22:46 +0200 Subject: tls: Fix TLS ulp context leak, when TLS_TX setsockopt is not used. Previously the TLS ulp context would leak if we attached a TLS ulp to a socket but did not use the TLS_TX setsockopt, or did use it but it failed. This patch solves the issue by overriding prot[TLS_BASE_TX].close and fixing tls_sk_proto_close to work properly when its called with ctx->tx_conf == TLS_BASE_TX. This patch also removes ctx->free_resources as we can use ctx->tx_conf to obtain the relevant information. Fixes: 3c4d7559159b ('tls: kernel TLS support') Signed-off-by: Ilya Lesokhin Signed-off-by: David S. Miller --- include/net/tls.h | 2 +- net/tls/tls_main.c | 22 ++++++++++++++-------- net/tls/tls_sw.c | 4 ++-- 3 files changed, 17 insertions(+), 11 deletions(-) (limited to 'include/net') diff --git a/include/net/tls.h b/include/net/tls.h index f058a6e08eaa..7cb58a6b8fd0 100644 --- a/include/net/tls.h +++ b/include/net/tls.h @@ -99,7 +99,6 @@ struct tls_context { u16 pending_open_record_frags; int (*push_pending_record)(struct sock *sk, int flags); - void (*free_resources)(struct sock *sk); void (*sk_write_space)(struct sock *sk); void (*sk_proto_close)(struct sock *sk, long timeout); @@ -124,6 +123,7 @@ int tls_sw_sendmsg(struct sock *sk, struct msghdr *msg, size_t size); int tls_sw_sendpage(struct sock *sk, struct page *page, int offset, size_t size, int flags); void tls_sw_close(struct sock *sk, long timeout); +void tls_sw_free_tx_resources(struct sock *sk); void tls_sk_destruct(struct sock *sk, struct tls_context *ctx); void tls_icsk_clean_acked(struct sock *sk); diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c index de6a1416bc41..13427ee7c582 100644 --- a/net/tls/tls_main.c +++ b/net/tls/tls_main.c @@ -226,6 +226,12 @@ static void tls_sk_proto_close(struct sock *sk, long timeout) void (*sk_proto_close)(struct sock *sk, long timeout); lock_sock(sk); + sk_proto_close = ctx->sk_proto_close; + + if (ctx->tx_conf == TLS_BASE_TX) { + kfree(ctx); + goto skip_tx_cleanup; + } if (!tls_complete_pending_work(sk, ctx, 0, &timeo)) tls_handle_open_record(sk, 0); @@ -242,13 +248,14 @@ static void tls_sk_proto_close(struct sock *sk, long timeout) sg++; } } - ctx->free_resources(sk); + kfree(ctx->rec_seq); kfree(ctx->iv); - sk_proto_close = ctx->sk_proto_close; - kfree(ctx); + if (ctx->tx_conf == TLS_SW_TX) + tls_sw_free_tx_resources(sk); +skip_tx_cleanup: release_sock(sk); sk_proto_close(sk, timeout); } @@ -402,8 +409,6 @@ static int do_tls_setsockopt_tx(struct sock *sk, char __user *optval, ctx->sk_write_space = sk->sk_write_space; sk->sk_write_space = tls_write_space; - ctx->sk_proto_close = sk->sk_prot->close; - /* currently SW is default, we will have ethtool in future */ rc = tls_set_sw_offload(sk, ctx); tx_conf = TLS_SW_TX; @@ -464,6 +469,7 @@ static int tls_init(struct sock *sk) icsk->icsk_ulp_data = ctx; ctx->setsockopt = sk->sk_prot->setsockopt; ctx->getsockopt = sk->sk_prot->getsockopt; + ctx->sk_proto_close = sk->sk_prot->close; ctx->tx_conf = TLS_BASE_TX; update_sk_prot(sk, ctx); @@ -480,11 +486,11 @@ static struct tcp_ulp_ops tcp_tls_ulp_ops __read_mostly = { static void build_protos(struct proto *prot, struct proto *base) { prot[TLS_BASE_TX] = *base; - prot[TLS_BASE_TX].setsockopt = tls_setsockopt; - prot[TLS_BASE_TX].getsockopt = tls_getsockopt; + prot[TLS_BASE_TX].setsockopt = tls_setsockopt; + prot[TLS_BASE_TX].getsockopt = tls_getsockopt; + prot[TLS_BASE_TX].close = tls_sk_proto_close; prot[TLS_SW_TX] = prot[TLS_BASE_TX]; - prot[TLS_SW_TX].close = tls_sk_proto_close; prot[TLS_SW_TX].sendmsg = tls_sw_sendmsg; prot[TLS_SW_TX].sendpage = tls_sw_sendpage; } diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index f00383a37622..fcd92a9c2d06 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -639,7 +639,7 @@ sendpage_end: return ret; } -static void tls_sw_free_resources(struct sock *sk) +void tls_sw_free_tx_resources(struct sock *sk) { struct tls_context *tls_ctx = tls_get_ctx(sk); struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx); @@ -650,6 +650,7 @@ static void tls_sw_free_resources(struct sock *sk) tls_free_both_sg(sk); kfree(ctx); + kfree(tls_ctx); } int tls_set_sw_offload(struct sock *sk, struct tls_context *ctx) @@ -679,7 +680,6 @@ int tls_set_sw_offload(struct sock *sk, struct tls_context *ctx) } ctx->priv_ctx = (struct tls_offload_context *)sw_ctx; - ctx->free_resources = tls_sw_free_resources; crypto_info = &ctx->crypto_send; switch (crypto_info->cipher_type) { -- cgit v1.2.3 From 213ef6e7c9c063c482d77f12cc438872628d48ec Mon Sep 17 00:00:00 2001 From: Ilya Lesokhin Date: Mon, 13 Nov 2017 10:22:47 +0200 Subject: tls: Move tls_make_aad to header to allow sharing move tls_make_aad as it is going to be reused by the device offload code and rx path. Remove unused recv parameter. Signed-off-by: Ilya Lesokhin Signed-off-by: David S. Miller --- include/net/tls.h | 15 +++++++++++++++ net/tls/tls_sw.c | 18 +----------------- 2 files changed, 16 insertions(+), 17 deletions(-) (limited to 'include/net') diff --git a/include/net/tls.h b/include/net/tls.h index 7cb58a6b8fd0..70becd0a9299 100644 --- a/include/net/tls.h +++ b/include/net/tls.h @@ -214,6 +214,21 @@ static inline void tls_fill_prepend(struct tls_context *ctx, ctx->iv + TLS_CIPHER_AES_GCM_128_SALT_SIZE, iv_size); } +static inline void tls_make_aad(char *buf, + size_t size, + char *record_sequence, + int record_sequence_size, + unsigned char record_type) +{ + memcpy(buf, record_sequence, record_sequence_size); + + buf[8] = record_type; + buf[9] = TLS_1_2_VERSION_MAJOR; + buf[10] = TLS_1_2_VERSION_MINOR; + buf[11] = size >> 8; + buf[12] = size & 0xFF; +} + static inline struct tls_context *tls_get_ctx(const struct sock *sk) { struct inet_connection_sock *icsk = inet_csk(sk); diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index fcd92a9c2d06..73d19210dd49 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -39,22 +39,6 @@ #include -static inline void tls_make_aad(int recv, - char *buf, - size_t size, - char *record_sequence, - int record_sequence_size, - unsigned char record_type) -{ - memcpy(buf, record_sequence, record_sequence_size); - - buf[8] = record_type; - buf[9] = TLS_1_2_VERSION_MAJOR; - buf[10] = TLS_1_2_VERSION_MINOR; - buf[11] = size >> 8; - buf[12] = size & 0xFF; -} - static void trim_sg(struct sock *sk, struct scatterlist *sg, int *sg_num_elem, unsigned int *sg_size, int target_size) { @@ -249,7 +233,7 @@ static int tls_push_record(struct sock *sk, int flags, sg_mark_end(ctx->sg_plaintext_data + ctx->sg_plaintext_num_elem - 1); sg_mark_end(ctx->sg_encrypted_data + ctx->sg_encrypted_num_elem - 1); - tls_make_aad(0, ctx->aad_space, ctx->sg_plaintext_size, + tls_make_aad(ctx->aad_space, ctx->sg_plaintext_size, tls_ctx->rec_seq, tls_ctx->rec_seq_size, record_type); -- cgit v1.2.3 From b9f3eb499d84f8d4adcb2f9212ec655700b28228 Mon Sep 17 00:00:00 2001 From: "Dmitry V. Levin" Date: Tue, 14 Nov 2017 06:30:11 +0300 Subject: uapi: fix linux/tls.h userspace compilation error Move inclusion of a private kernel header from uapi/linux/tls.h to its only user - net/tls.h, to fix the following linux/tls.h userspace compilation error: /usr/include/linux/tls.h:41:21: fatal error: net/tcp.h: No such file or directory As to this point uapi/linux/tls.h was totaly unusuable for userspace, cleanup this header file further by moving other redundant includes to net/tls.h. Fixes: 3c4d7559159b ("tls: kernel TLS support") Cc: # v4.13+ Signed-off-by: Dmitry V. Levin Signed-off-by: David S. Miller --- include/net/tls.h | 4 ++++ include/uapi/linux/tls.h | 4 ---- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include/net') diff --git a/include/net/tls.h b/include/net/tls.h index 70becd0a9299..936cfc5cab7d 100644 --- a/include/net/tls.h +++ b/include/net/tls.h @@ -35,6 +35,10 @@ #define _TLS_OFFLOAD_H #include +#include +#include +#include +#include #include diff --git a/include/uapi/linux/tls.h b/include/uapi/linux/tls.h index d5e0682ab837..293b2cdad88d 100644 --- a/include/uapi/linux/tls.h +++ b/include/uapi/linux/tls.h @@ -35,10 +35,6 @@ #define _UAPI_LINUX_TLS_H #include -#include -#include -#include -#include /* TLS socket options */ #define TLS_TX 1 /* Set transmit parameters */ -- cgit v1.2.3 From 6670e152447732ba90626f36dfc015a13fbf150e Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 14 Nov 2017 08:25:49 -0800 Subject: tcp: Namespace-ify sysctl_tcp_default_congestion_control Make default TCP default congestion control to a per namespace value. This changes default congestion control to a pointer to congestion ops (rather than implicit as first element of available lsit). The congestion control setting of new namespaces is inherited from the current setting of the root namespace. Signed-off-by: Stephen Hemminger Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + include/net/tcp.h | 6 ++-- net/ipv4/fib_semantics.c | 4 +-- net/ipv4/sysctl_net_ipv4.c | 19 +++++++----- net/ipv4/tcp_cong.c | 76 ++++++++++++++++++++++------------------------ net/ipv4/tcp_ipv4.c | 9 ++++++ net/ipv6/route.c | 3 +- 7 files changed, 64 insertions(+), 54 deletions(-) (limited to 'include/net') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 5e12975fc658..44668c29701a 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -160,6 +160,7 @@ struct netns_ipv4 { struct inet_timewait_death_row tcp_death_row; int sysctl_max_syn_backlog; int sysctl_tcp_fastopen; + const struct tcp_congestion_ops __rcu *tcp_congestion_control; struct tcp_fastopen_context __rcu *tcp_fastopen_ctx; spinlock_t tcp_fastopen_ctx_lock; unsigned int sysctl_tcp_fastopen_blackhole_timeout; diff --git a/include/net/tcp.h b/include/net/tcp.h index ed71511e67a6..35cc7d0d3d47 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1002,8 +1002,8 @@ void tcp_unregister_congestion_control(struct tcp_congestion_ops *type); void tcp_assign_congestion_control(struct sock *sk); void tcp_init_congestion_control(struct sock *sk); void tcp_cleanup_congestion_control(struct sock *sk); -int tcp_set_default_congestion_control(const char *name); -void tcp_get_default_congestion_control(char *name); +int tcp_set_default_congestion_control(struct net *net, const char *name); +void tcp_get_default_congestion_control(struct net *net, char *name); void tcp_get_available_congestion_control(char *buf, size_t len); void tcp_get_allowed_congestion_control(char *buf, size_t len); int tcp_set_allowed_congestion_control(char *allowed); @@ -1017,7 +1017,7 @@ void tcp_reno_cong_avoid(struct sock *sk, u32 ack, u32 acked); extern struct tcp_congestion_ops tcp_reno; struct tcp_congestion_ops *tcp_ca_find_key(u32 key); -u32 tcp_ca_get_key_by_name(const char *name, bool *ecn_ca); +u32 tcp_ca_get_key_by_name(struct net *net, const char *name, bool *ecn_ca); #ifdef CONFIG_INET char *tcp_ca_get_name_by_key(u32 key, char *buffer); #else diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 589caaa90613..f04d944f8abe 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -710,7 +710,7 @@ bool fib_metrics_match(struct fib_config *cfg, struct fib_info *fi) bool ecn_ca = false; nla_strlcpy(tmp, nla, sizeof(tmp)); - val = tcp_ca_get_key_by_name(tmp, &ecn_ca); + val = tcp_ca_get_key_by_name(fi->fib_net, tmp, &ecn_ca); } else { val = nla_get_u32(nla); } @@ -1030,7 +1030,7 @@ fib_convert_metrics(struct fib_info *fi, const struct fib_config *cfg) char tmp[TCP_CA_NAME_MAX]; nla_strlcpy(tmp, nla, sizeof(tmp)); - val = tcp_ca_get_key_by_name(tmp, &ecn_ca); + val = tcp_ca_get_key_by_name(fi->fib_net, tmp, &ecn_ca); if (val == TCP_CA_UNSPEC) return -EINVAL; } else { diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index ef0ff3357a44..93e172118a94 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -201,6 +201,8 @@ static int ipv4_ping_group_range(struct ctl_table *table, int write, static int proc_tcp_congestion_control(struct ctl_table *ctl, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { + struct net *net = container_of(ctl->data, struct net, + ipv4.tcp_congestion_control); char val[TCP_CA_NAME_MAX]; struct ctl_table tbl = { .data = val, @@ -208,11 +210,11 @@ static int proc_tcp_congestion_control(struct ctl_table *ctl, int write, }; int ret; - tcp_get_default_congestion_control(val); + tcp_get_default_congestion_control(net, val); ret = proc_dostring(&tbl, write, buffer, lenp, ppos); if (write && ret == 0) - ret = tcp_set_default_congestion_control(val); + ret = tcp_set_default_congestion_control(net, val); return ret; } @@ -447,12 +449,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { - .procname = "tcp_congestion_control", - .mode = 0644, - .maxlen = TCP_CA_NAME_MAX, - .proc_handler = proc_tcp_congestion_control, - }, #ifdef CONFIG_NETLABEL { .procname = "cipso_cache_enable", @@ -763,6 +759,13 @@ static struct ctl_table ipv4_net_table[] = { .extra1 = &one }, #endif + { + .procname = "tcp_congestion_control", + .data = &init_net.ipv4.tcp_congestion_control, + .mode = 0644, + .maxlen = TCP_CA_NAME_MAX, + .proc_handler = proc_tcp_congestion_control, + }, { .procname = "tcp_keepalive_time", .data = &init_net.ipv4.sysctl_tcp_keepalive_time, diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index 2f26124fd160..bc6c02f16243 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -33,9 +33,11 @@ static struct tcp_congestion_ops *tcp_ca_find(const char *name) } /* Must be called with rcu lock held */ -static const struct tcp_congestion_ops *__tcp_ca_find_autoload(const char *name) +static struct tcp_congestion_ops *tcp_ca_find_autoload(struct net *net, + const char *name) { - const struct tcp_congestion_ops *ca = tcp_ca_find(name); + struct tcp_congestion_ops *ca = tcp_ca_find(name); + #ifdef CONFIG_MODULES if (!ca && capable(CAP_NET_ADMIN)) { rcu_read_unlock(); @@ -115,7 +117,7 @@ void tcp_unregister_congestion_control(struct tcp_congestion_ops *ca) } EXPORT_SYMBOL_GPL(tcp_unregister_congestion_control); -u32 tcp_ca_get_key_by_name(const char *name, bool *ecn_ca) +u32 tcp_ca_get_key_by_name(struct net *net, const char *name, bool *ecn_ca) { const struct tcp_congestion_ops *ca; u32 key = TCP_CA_UNSPEC; @@ -123,7 +125,7 @@ u32 tcp_ca_get_key_by_name(const char *name, bool *ecn_ca) might_sleep(); rcu_read_lock(); - ca = __tcp_ca_find_autoload(name); + ca = tcp_ca_find_autoload(net, name); if (ca) { key = ca->key; *ecn_ca = ca->flags & TCP_CONG_NEEDS_ECN; @@ -153,23 +155,18 @@ EXPORT_SYMBOL_GPL(tcp_ca_get_name_by_key); /* Assign choice of congestion control. */ void tcp_assign_congestion_control(struct sock *sk) { + struct net *net = sock_net(sk); struct inet_connection_sock *icsk = inet_csk(sk); - struct tcp_congestion_ops *ca; + const struct tcp_congestion_ops *ca; rcu_read_lock(); - list_for_each_entry_rcu(ca, &tcp_cong_list, list) { - if (likely(try_module_get(ca->owner))) { - icsk->icsk_ca_ops = ca; - goto out; - } - /* Fallback to next available. The last really - * guaranteed fallback is Reno from this list. - */ - } -out: + ca = rcu_dereference(net->ipv4.tcp_congestion_control); + if (unlikely(!try_module_get(ca->owner))) + ca = &tcp_reno; + icsk->icsk_ca_ops = ca; rcu_read_unlock(); - memset(icsk->icsk_ca_priv, 0, sizeof(icsk->icsk_ca_priv)); + memset(icsk->icsk_ca_priv, 0, sizeof(icsk->icsk_ca_priv)); if (ca->flags & TCP_CONG_NEEDS_ECN) INET_ECN_xmit(sk); else @@ -214,29 +211,27 @@ void tcp_cleanup_congestion_control(struct sock *sk) } /* Used by sysctl to change default congestion control */ -int tcp_set_default_congestion_control(const char *name) +int tcp_set_default_congestion_control(struct net *net, const char *name) { struct tcp_congestion_ops *ca; - int ret = -ENOENT; - - spin_lock(&tcp_cong_list_lock); - ca = tcp_ca_find(name); -#ifdef CONFIG_MODULES - if (!ca && capable(CAP_NET_ADMIN)) { - spin_unlock(&tcp_cong_list_lock); + const struct tcp_congestion_ops *prev; + int ret; - request_module("tcp_%s", name); - spin_lock(&tcp_cong_list_lock); - ca = tcp_ca_find(name); - } -#endif + rcu_read_lock(); + ca = tcp_ca_find_autoload(net, name); + if (!ca) { + ret = -ENOENT; + } else if (!try_module_get(ca->owner)) { + ret = -EBUSY; + } else { + prev = xchg(&net->ipv4.tcp_congestion_control, ca); + if (prev) + module_put(prev->owner); - if (ca) { - ca->flags |= TCP_CONG_NON_RESTRICTED; /* default is always allowed */ - list_move(&ca->list, &tcp_cong_list); + ca->flags |= TCP_CONG_NON_RESTRICTED; ret = 0; } - spin_unlock(&tcp_cong_list_lock); + rcu_read_unlock(); return ret; } @@ -244,7 +239,8 @@ int tcp_set_default_congestion_control(const char *name) /* Set default value from kernel configuration at bootup */ static int __init tcp_congestion_default(void) { - return tcp_set_default_congestion_control(CONFIG_DEFAULT_TCP_CONG); + return tcp_set_default_congestion_control(&init_net, + CONFIG_DEFAULT_TCP_CONG); } late_initcall(tcp_congestion_default); @@ -264,14 +260,12 @@ void tcp_get_available_congestion_control(char *buf, size_t maxlen) } /* Get current default congestion control */ -void tcp_get_default_congestion_control(char *name) +void tcp_get_default_congestion_control(struct net *net, char *name) { - struct tcp_congestion_ops *ca; - /* We will always have reno... */ - BUG_ON(list_empty(&tcp_cong_list)); + const struct tcp_congestion_ops *ca; rcu_read_lock(); - ca = list_entry(tcp_cong_list.next, struct tcp_congestion_ops, list); + ca = rcu_dereference(net->ipv4.tcp_congestion_control); strncpy(name, ca->name, TCP_CA_NAME_MAX); rcu_read_unlock(); } @@ -351,12 +345,14 @@ int tcp_set_congestion_control(struct sock *sk, const char *name, bool load, boo if (!load) ca = tcp_ca_find(name); else - ca = __tcp_ca_find_autoload(name); + ca = tcp_ca_find_autoload(sock_net(sk), name); + /* No change asking for existing value */ if (ca == icsk->icsk_ca_ops) { icsk->icsk_ca_setsockopt = 1; goto out; } + if (!ca) { err = -ENOENT; } else if (!load) { diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 1eac84b8044e..c6bc0c4d19c6 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2430,6 +2430,8 @@ static void __net_exit tcp_sk_exit(struct net *net) { int cpu; + module_put(net->ipv4.tcp_congestion_control->owner); + for_each_possible_cpu(cpu) inet_ctl_sock_destroy(*per_cpu_ptr(net->ipv4.tcp_sk, cpu)); free_percpu(net->ipv4.tcp_sk); @@ -2522,6 +2524,13 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_fastopen_blackhole_timeout = 60 * 60; atomic_set(&net->ipv4.tfo_active_disable_times, 0); + /* Reno is always built in */ + if (!net_eq(net, &init_net) && + try_module_get(init_net.ipv4.tcp_congestion_control->owner)) + net->ipv4.tcp_congestion_control = init_net.ipv4.tcp_congestion_control; + else + net->ipv4.tcp_congestion_control = &tcp_reno; + return 0; fail: tcp_sk_exit(net); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 70d9659fc1e9..05eb7bc36156 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2378,6 +2378,7 @@ out: static int ip6_convert_metrics(struct mx6_config *mxc, const struct fib6_config *cfg) { + struct net *net = cfg->fc_nlinfo.nl_net; bool ecn_ca = false; struct nlattr *nla; int remaining; @@ -2403,7 +2404,7 @@ static int ip6_convert_metrics(struct mx6_config *mxc, char tmp[TCP_CA_NAME_MAX]; nla_strlcpy(tmp, nla, sizeof(tmp)); - val = tcp_ca_get_key_by_name(tmp, &ecn_ca); + val = tcp_ca_get_key_by_name(net, tmp, &ecn_ca); if (val == TCP_CA_UNSPEC) goto err; } else { -- cgit v1.2.3 From 50895b9de1d3e0258e015e8e55128d835d9a9f19 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 14 Nov 2017 21:02:19 -0800 Subject: tcp: highest_sack fix syzbot easily found a regression added in our latest patches [1] No longer set tp->highest_sack to the head of the send queue since this is not logical and error prone. Only sack processing should maintain the pointer to an skb from rtx queue. We might in the future only remember the sequence instead of a pointer to skb, since rb-tree should allow a fast lookup. [1] BUG: KASAN: use-after-free in tcp_highest_sack_seq include/net/tcp.h:1706 [inline] BUG: KASAN: use-after-free in tcp_ack+0x42bb/0x4fd0 net/ipv4/tcp_input.c:3537 Read of size 4 at addr ffff8801c154faa8 by task syz-executor4/12860 CPU: 0 PID: 12860 Comm: syz-executor4 Not tainted 4.14.0-next-20171113+ #41 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Call Trace: __dump_stack lib/dump_stack.c:17 [inline] dump_stack+0x194/0x257 lib/dump_stack.c:53 print_address_description+0x73/0x250 mm/kasan/report.c:252 kasan_report_error mm/kasan/report.c:351 [inline] kasan_report+0x25b/0x340 mm/kasan/report.c:409 __asan_report_load4_noabort+0x14/0x20 mm/kasan/report.c:429 tcp_highest_sack_seq include/net/tcp.h:1706 [inline] tcp_ack+0x42bb/0x4fd0 net/ipv4/tcp_input.c:3537 tcp_rcv_established+0x672/0x18a0 net/ipv4/tcp_input.c:5439 tcp_v4_do_rcv+0x2ab/0x7d0 net/ipv4/tcp_ipv4.c:1468 sk_backlog_rcv include/net/sock.h:909 [inline] __release_sock+0x124/0x360 net/core/sock.c:2264 release_sock+0xa4/0x2a0 net/core/sock.c:2778 tcp_sendmsg+0x3a/0x50 net/ipv4/tcp.c:1462 inet_sendmsg+0x11f/0x5e0 net/ipv4/af_inet.c:763 sock_sendmsg_nosec net/socket.c:632 [inline] sock_sendmsg+0xca/0x110 net/socket.c:642 ___sys_sendmsg+0x75b/0x8a0 net/socket.c:2048 __sys_sendmsg+0xe5/0x210 net/socket.c:2082 SYSC_sendmsg net/socket.c:2093 [inline] SyS_sendmsg+0x2d/0x50 net/socket.c:2089 entry_SYSCALL_64_fastpath+0x1f/0x96 RIP: 0033:0x452879 RSP: 002b:00007fc9761bfbe8 EFLAGS: 00000212 ORIG_RAX: 000000000000002e RAX: ffffffffffffffda RBX: 0000000000758020 RCX: 0000000000452879 RDX: 0000000000000000 RSI: 0000000020917fc8 RDI: 0000000000000015 RBP: 0000000000000086 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000212 R12: 00000000006ee3a0 R13: 00000000ffffffff R14: 00007fc9761c06d4 R15: 0000000000000000 Allocated by task 12860: save_stack+0x43/0xd0 mm/kasan/kasan.c:447 set_track mm/kasan/kasan.c:459 [inline] kasan_kmalloc+0xad/0xe0 mm/kasan/kasan.c:551 kasan_slab_alloc+0x12/0x20 mm/kasan/kasan.c:489 kmem_cache_alloc_node+0x144/0x760 mm/slab.c:3638 __alloc_skb+0xf1/0x780 net/core/skbuff.c:193 alloc_skb_fclone include/linux/skbuff.h:1023 [inline] sk_stream_alloc_skb+0x11d/0x900 net/ipv4/tcp.c:870 tcp_sendmsg_locked+0x1341/0x3b80 net/ipv4/tcp.c:1299 tcp_sendmsg+0x2f/0x50 net/ipv4/tcp.c:1461 inet_sendmsg+0x11f/0x5e0 net/ipv4/af_inet.c:763 sock_sendmsg_nosec net/socket.c:632 [inline] sock_sendmsg+0xca/0x110 net/socket.c:642 SYSC_sendto+0x358/0x5a0 net/socket.c:1749 SyS_sendto+0x40/0x50 net/socket.c:1717 entry_SYSCALL_64_fastpath+0x1f/0x96 Freed by task 12860: save_stack+0x43/0xd0 mm/kasan/kasan.c:447 set_track mm/kasan/kasan.c:459 [inline] kasan_slab_free+0x71/0xc0 mm/kasan/kasan.c:524 __cache_free mm/slab.c:3492 [inline] kmem_cache_free+0x77/0x280 mm/slab.c:3750 kfree_skbmem+0xdd/0x1d0 net/core/skbuff.c:603 __kfree_skb+0x1d/0x20 net/core/skbuff.c:642 sk_wmem_free_skb include/net/sock.h:1419 [inline] tcp_rtx_queue_unlink_and_free include/net/tcp.h:1682 [inline] tcp_clean_rtx_queue net/ipv4/tcp_input.c:3111 [inline] tcp_ack+0x1b17/0x4fd0 net/ipv4/tcp_input.c:3593 tcp_rcv_established+0x672/0x18a0 net/ipv4/tcp_input.c:5439 tcp_v4_do_rcv+0x2ab/0x7d0 net/ipv4/tcp_ipv4.c:1468 sk_backlog_rcv include/net/sock.h:909 [inline] __release_sock+0x124/0x360 net/core/sock.c:2264 release_sock+0xa4/0x2a0 net/core/sock.c:2778 tcp_sendmsg+0x3a/0x50 net/ipv4/tcp.c:1462 inet_sendmsg+0x11f/0x5e0 net/ipv4/af_inet.c:763 sock_sendmsg_nosec net/socket.c:632 [inline] sock_sendmsg+0xca/0x110 net/socket.c:642 ___sys_sendmsg+0x75b/0x8a0 net/socket.c:2048 __sys_sendmsg+0xe5/0x210 net/socket.c:2082 SYSC_sendmsg net/socket.c:2093 [inline] SyS_sendmsg+0x2d/0x50 net/socket.c:2089 entry_SYSCALL_64_fastpath+0x1f/0x96 The buggy address belongs to the object at ffff8801c154fa80 which belongs to the cache skbuff_fclone_cache of size 456 The buggy address is located 40 bytes inside of 456-byte region [ffff8801c154fa80, ffff8801c154fc48) The buggy address belongs to the page: page:ffffea00070553c0 count:1 mapcount:0 mapping:ffff8801c154f080 index:0x0 flags: 0x2fffc0000000100(slab) raw: 02fffc0000000100 ffff8801c154f080 0000000000000000 0000000100000006 raw: ffffea00070a5a20 ffffea0006a18360 ffff8801d9ca0500 0000000000000000 page dumped because: kasan: bad access detected Fixes: 737ff314563c ("tcp: use sequence distance to detect reordering") Signed-off-by: Eric Dumazet Cc: Yuchung Cheng Reported-by: syzbot Signed-off-by: David S. Miller --- include/net/tcp.h | 17 +++-------------- net/ipv4/tcp_input.c | 2 +- 2 files changed, 4 insertions(+), 15 deletions(-) (limited to 'include/net') diff --git a/include/net/tcp.h b/include/net/tcp.h index 35cc7d0d3d47..85ea578195d4 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1630,9 +1630,6 @@ static inline void tcp_check_send_head(struct sock *sk, struct sk_buff *skb_unli { if (tcp_write_queue_empty(sk)) tcp_chrono_stop(sk, TCP_CHRONO_BUSY); - - if (tcp_sk(sk)->highest_sack == skb_unlinked) - tcp_sk(sk)->highest_sack = NULL; } static inline void __tcp_add_write_queue_tail(struct sock *sk, struct sk_buff *skb) @@ -1645,12 +1642,8 @@ static inline void tcp_add_write_queue_tail(struct sock *sk, struct sk_buff *skb __tcp_add_write_queue_tail(sk, skb); /* Queue it, remembering where we must start sending. */ - if (sk->sk_write_queue.next == skb) { + if (sk->sk_write_queue.next == skb) tcp_chrono_start(sk, TCP_CHRONO_BUSY); - - if (tcp_sk(sk)->highest_sack == NULL) - tcp_sk(sk)->highest_sack = skb; - } } /* Insert new before skb on the write queue of sk. */ @@ -1708,9 +1701,7 @@ static inline u32 tcp_highest_sack_seq(struct tcp_sock *tp) static inline void tcp_advance_highest_sack(struct sock *sk, struct sk_buff *skb) { - struct sk_buff *next = skb_rb_next(skb); - - tcp_sk(sk)->highest_sack = next ?: tcp_send_head(sk); + tcp_sk(sk)->highest_sack = skb_rb_next(skb); } static inline struct sk_buff *tcp_highest_sack(struct sock *sk) @@ -1720,9 +1711,7 @@ static inline struct sk_buff *tcp_highest_sack(struct sock *sk) static inline void tcp_highest_sack_reset(struct sock *sk) { - struct sk_buff *skb = tcp_rtx_queue_head(sk); - - tcp_sk(sk)->highest_sack = skb ?: tcp_send_head(sk); + tcp_sk(sk)->highest_sack = tcp_rtx_queue_head(sk); } /* Called when old skb is about to be deleted and replaced by new skb */ diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index c3447c5512fd..f0b572fe959a 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -3534,7 +3534,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) icsk->icsk_retransmits = 0; } - prior_fack = tcp_highest_sack_seq(tp); + prior_fack = tcp_is_sack(tp) ? tcp_highest_sack_seq(tp) : tp->snd_una; rs.prior_in_flight = tcp_packets_in_flight(tp); /* ts_recent update must be made after we are sure that the packet -- cgit v1.2.3