diff options
Diffstat (limited to 'net')
42 files changed, 462 insertions, 197 deletions
diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c index ead18ca836de..11a23fd6e1a0 100644 --- a/net/batman-adv/fragmentation.c +++ b/net/batman-adv/fragmentation.c @@ -239,8 +239,10 @@ err_unlock: spin_unlock_bh(&chain->lock); err: - if (!ret) + if (!ret) { kfree(frag_entry_new); + kfree_skb(skb); + } return ret; } @@ -313,7 +315,7 @@ free: * * There are three possible outcomes: 1) Packet is merged: Return true and * set *skb to merged packet; 2) Packet is buffered: Return true and set *skb - * to NULL; 3) Error: Return false and leave skb as is. + * to NULL; 3) Error: Return false and free skb. * * Return: true when packet is merged or buffered, false when skb is not not * used. @@ -338,9 +340,9 @@ bool batadv_frag_skb_buffer(struct sk_buff **skb, goto out_err; out: - *skb = skb_out; ret = true; out_err: + *skb = skb_out; return ret; } @@ -499,6 +501,12 @@ int batadv_frag_send_packet(struct sk_buff *skb, /* Eat and send fragments from the tail of skb */ while (skb->len > max_fragment_size) { + /* The initial check in this function should cover this case */ + if (unlikely(frag_header.no == BATADV_FRAG_MAX_FRAGMENTS - 1)) { + ret = -EINVAL; + goto put_primary_if; + } + skb_fragment = batadv_frag_create(skb, &frag_header, mtu); if (!skb_fragment) { ret = -ENOMEM; @@ -515,12 +523,6 @@ int batadv_frag_send_packet(struct sk_buff *skb, } frag_header.no++; - - /* The initial check in this function should cover this case */ - if (frag_header.no == BATADV_FRAG_MAX_FRAGMENTS - 1) { - ret = -EINVAL; - goto put_primary_if; - } } /* Make room for the fragment header. */ diff --git a/net/batman-adv/types.h b/net/batman-adv/types.h index 8f64a5c01345..66b25e410a41 100644 --- a/net/batman-adv/types.h +++ b/net/batman-adv/types.h @@ -402,7 +402,7 @@ struct batadv_gw_node { struct rcu_head rcu; }; -DECLARE_EWMA(throughput, 1024, 8) +DECLARE_EWMA(throughput, 10, 8) /** * struct batadv_hardif_neigh_node_bat_v - B.A.T.M.A.N. V private neighbor diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index 6bfac29318f2..902af6ba481c 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -186,8 +186,9 @@ void br_flood(struct net_bridge *br, struct sk_buff *skb, /* Do not flood unicast traffic to ports that turn it off */ if (pkt_type == BR_PKT_UNICAST && !(p->flags & BR_FLOOD)) continue; + /* Do not flood if mc off, except for traffic we originate */ if (pkt_type == BR_PKT_MULTICAST && - !(p->flags & BR_MCAST_FLOOD)) + !(p->flags & BR_MCAST_FLOOD) && skb->dev != br->dev) continue; /* Do not flood to ports that enable proxy ARP */ diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c index 62e68c0dc687..b838213c408e 100644 --- a/net/bridge/br_vlan.c +++ b/net/bridge/br_vlan.c @@ -997,10 +997,10 @@ err_vlan_add: RCU_INIT_POINTER(p->vlgrp, NULL); synchronize_rcu(); vlan_tunnel_deinit(vg); -err_vlan_enabled: err_tunnel_init: rhashtable_destroy(&vg->vlan_hash); err_rhtbl: +err_vlan_enabled: kfree(vg); goto out; diff --git a/net/core/dev.c b/net/core/dev.c index 304f2deae5f9..8637b2b71f3d 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1698,27 +1698,54 @@ EXPORT_SYMBOL_GPL(net_dec_egress_queue); static struct static_key netstamp_needed __read_mostly; #ifdef HAVE_JUMP_LABEL static atomic_t netstamp_needed_deferred; +static atomic_t netstamp_wanted; static void netstamp_clear(struct work_struct *work) { int deferred = atomic_xchg(&netstamp_needed_deferred, 0); + int wanted; - while (deferred--) - static_key_slow_dec(&netstamp_needed); + wanted = atomic_add_return(deferred, &netstamp_wanted); + if (wanted > 0) + static_key_enable(&netstamp_needed); + else + static_key_disable(&netstamp_needed); } static DECLARE_WORK(netstamp_work, netstamp_clear); #endif void net_enable_timestamp(void) { +#ifdef HAVE_JUMP_LABEL + int wanted; + + while (1) { + wanted = atomic_read(&netstamp_wanted); + if (wanted <= 0) + break; + if (atomic_cmpxchg(&netstamp_wanted, wanted, wanted + 1) == wanted) + return; + } + atomic_inc(&netstamp_needed_deferred); + schedule_work(&netstamp_work); +#else static_key_slow_inc(&netstamp_needed); +#endif } EXPORT_SYMBOL(net_enable_timestamp); void net_disable_timestamp(void) { #ifdef HAVE_JUMP_LABEL - /* net_disable_timestamp() can be called from non process context */ - atomic_inc(&netstamp_needed_deferred); + int wanted; + + while (1) { + wanted = atomic_read(&netstamp_wanted); + if (wanted <= 1) + break; + if (atomic_cmpxchg(&netstamp_wanted, wanted, wanted - 1) == wanted) + return; + } + atomic_dec(&netstamp_needed_deferred); schedule_work(&netstamp_work); #else static_key_slow_dec(&netstamp_needed); @@ -4884,6 +4911,39 @@ void __napi_schedule(struct napi_struct *n) EXPORT_SYMBOL(__napi_schedule); /** + * napi_schedule_prep - check if napi can be scheduled + * @n: napi context + * + * Test if NAPI routine is already running, and if not mark + * it as running. This is used as a condition variable + * insure only one NAPI poll instance runs. We also make + * sure there is no pending NAPI disable. + */ +bool napi_schedule_prep(struct napi_struct *n) +{ + unsigned long val, new; + + do { + val = READ_ONCE(n->state); + if (unlikely(val & NAPIF_STATE_DISABLE)) + return false; + new = val | NAPIF_STATE_SCHED; + + /* Sets STATE_MISSED bit if STATE_SCHED was already set + * This was suggested by Alexander Duyck, as compiler + * emits better code than : + * if (val & NAPIF_STATE_SCHED) + * new |= NAPIF_STATE_MISSED; + */ + new |= (val & NAPIF_STATE_SCHED) / NAPIF_STATE_SCHED * + NAPIF_STATE_MISSED; + } while (cmpxchg(&n->state, val, new) != val); + + return !(val & NAPIF_STATE_SCHED); +} +EXPORT_SYMBOL(napi_schedule_prep); + +/** * __napi_schedule_irqoff - schedule for receive * @n: entry to schedule * @@ -4897,7 +4957,7 @@ EXPORT_SYMBOL(__napi_schedule_irqoff); bool napi_complete_done(struct napi_struct *n, int work_done) { - unsigned long flags; + unsigned long flags, val, new; /* * 1) Don't let napi dequeue from the cpu poll list @@ -4927,7 +4987,27 @@ bool napi_complete_done(struct napi_struct *n, int work_done) list_del_init(&n->poll_list); local_irq_restore(flags); } - WARN_ON_ONCE(!test_and_clear_bit(NAPI_STATE_SCHED, &n->state)); + + do { + val = READ_ONCE(n->state); + + WARN_ON_ONCE(!(val & NAPIF_STATE_SCHED)); + + new = val & ~(NAPIF_STATE_MISSED | NAPIF_STATE_SCHED); + + /* If STATE_MISSED was set, leave STATE_SCHED set, + * because we will call napi->poll() one more time. + * This C code was suggested by Alexander Duyck to help gcc. + */ + new |= (val & NAPIF_STATE_MISSED) / NAPIF_STATE_MISSED * + NAPIF_STATE_SCHED; + } while (cmpxchg(&n->state, val, new) != val); + + if (unlikely(val & NAPIF_STATE_MISSED)) { + __napi_schedule(n); + return false; + } + return true; } EXPORT_SYMBOL(napi_complete_done); @@ -4953,6 +5033,16 @@ static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock) { int rc; + /* Busy polling means there is a high chance device driver hard irq + * could not grab NAPI_STATE_SCHED, and that NAPI_STATE_MISSED was + * set in napi_schedule_prep(). + * Since we are about to call napi->poll() once more, we can safely + * clear NAPI_STATE_MISSED. + * + * Note: x86 could use a single "lock and ..." instruction + * to perform these two clear_bit() + */ + clear_bit(NAPI_STATE_MISSED, &napi->state); clear_bit(NAPI_STATE_IN_BUSY_POLL, &napi->state); local_bh_disable(); @@ -5088,8 +5178,13 @@ static enum hrtimer_restart napi_watchdog(struct hrtimer *timer) struct napi_struct *napi; napi = container_of(timer, struct napi_struct, timer); - if (napi->gro_list) - napi_schedule_irqoff(napi); + + /* Note : we use a relaxed variant of napi_schedule_prep() not setting + * NAPI_STATE_MISSED, since we do not react to a device IRQ. + */ + if (napi->gro_list && !napi_disable_pending(napi) && + !test_and_set_bit(NAPI_STATE_SCHED, &napi->state)) + __napi_schedule_irqoff(napi); return HRTIMER_NORESTART; } diff --git a/net/core/sock.c b/net/core/sock.c index e7d74940e863..f6fd79f33097 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1539,11 +1539,7 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority) is_charged = sk_filter_charge(newsk, filter); if (unlikely(!is_charged || xfrm_sk_clone_policy(newsk, sk))) { - /* It is still raw copy of parent, so invalidate - * destructor and make plain sk_free() */ - newsk->sk_destruct = NULL; - bh_unlock_sock(newsk); - sk_free(newsk); + sk_free_unlock_clone(newsk); newsk = NULL; goto out; } @@ -1592,6 +1588,16 @@ out: } EXPORT_SYMBOL_GPL(sk_clone_lock); +void sk_free_unlock_clone(struct sock *sk) +{ + /* It is still raw copy of parent, so invalidate + * destructor and make plain sk_free() */ + sk->sk_destruct = NULL; + bh_unlock_sock(sk); + sk_free(sk); +} +EXPORT_SYMBOL_GPL(sk_free_unlock_clone); + void sk_setup_caps(struct sock *sk, struct dst_entry *dst) { u32 max_segs = 1; diff --git a/net/dccp/input.c b/net/dccp/input.c index 8fedc2d49770..4a05d7876850 100644 --- a/net/dccp/input.c +++ b/net/dccp/input.c @@ -577,6 +577,7 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb, struct dccp_sock *dp = dccp_sk(sk); struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb); const int old_state = sk->sk_state; + bool acceptable; int queued = 0; /* @@ -603,8 +604,13 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb, */ if (sk->sk_state == DCCP_LISTEN) { if (dh->dccph_type == DCCP_PKT_REQUEST) { - if (inet_csk(sk)->icsk_af_ops->conn_request(sk, - skb) < 0) + /* It is possible that we process SYN packets from backlog, + * so we need to make sure to disable BH right there. + */ + local_bh_disable(); + acceptable = inet_csk(sk)->icsk_af_ops->conn_request(sk, skb) >= 0; + local_bh_enable(); + if (!acceptable) return 1; consume_skb(skb); return 0; diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c index 53eddf99e4f6..e267e6f4c9a5 100644 --- a/net/dccp/minisocks.c +++ b/net/dccp/minisocks.c @@ -119,10 +119,7 @@ struct sock *dccp_create_openreq_child(const struct sock *sk, * Activate features: initialise CCIDs, sequence windows etc. */ if (dccp_feat_activate_values(newsk, &dreq->dreq_featneg)) { - /* It is still raw copy of parent, so invalidate - * destructor and make plain sk_free() */ - newsk->sk_destruct = NULL; - sk_free(newsk); + sk_free_unlock_clone(newsk); return NULL; } dccp_init_xmit_timers(newsk); diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index b39a791f6756..42bfd08109dd 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -622,6 +622,7 @@ const struct nla_policy rtm_ipv4_policy[RTA_MAX + 1] = { [RTA_ENCAP_TYPE] = { .type = NLA_U16 }, [RTA_ENCAP] = { .type = NLA_NESTED }, [RTA_UID] = { .type = NLA_U32 }, + [RTA_MARK] = { .type = NLA_U32 }, }; static int rtm_to_fib_config(struct net *net, struct sk_buff *skb, diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c index b3cc1335adbc..c0cc6aa8cfaa 100644 --- a/net/ipv4/netfilter.c +++ b/net/ipv4/netfilter.c @@ -23,7 +23,8 @@ int ip_route_me_harder(struct net *net, struct sk_buff *skb, unsigned int addr_t struct rtable *rt; struct flowi4 fl4 = {}; __be32 saddr = iph->saddr; - __u8 flags = skb->sk ? inet_sk_flowi_flags(skb->sk) : 0; + const struct sock *sk = skb_to_full_sk(skb); + __u8 flags = sk ? inet_sk_flowi_flags(sk) : 0; struct net_device *dev = skb_dst(skb)->dev; unsigned int hh_len; @@ -40,7 +41,7 @@ int ip_route_me_harder(struct net *net, struct sk_buff *skb, unsigned int addr_t fl4.daddr = iph->daddr; fl4.saddr = saddr; fl4.flowi4_tos = RT_TOS(iph->tos); - fl4.flowi4_oif = skb->sk ? skb->sk->sk_bound_dev_if : 0; + fl4.flowi4_oif = sk ? sk->sk_bound_dev_if : 0; if (!fl4.flowi4_oif) fl4.flowi4_oif = l3mdev_master_ifindex(dev); fl4.flowi4_mark = skb->mark; @@ -61,7 +62,7 @@ int ip_route_me_harder(struct net *net, struct sk_buff *skb, unsigned int addr_t xfrm_decode_session(skb, flowi4_to_flowi(&fl4), AF_INET) == 0) { struct dst_entry *dst = skb_dst(skb); skb_dst_set(skb, NULL); - dst = xfrm_lookup(net, dst, flowi4_to_flowi(&fl4), skb->sk, 0); + dst = xfrm_lookup(net, dst, flowi4_to_flowi(&fl4), sk, 0); if (IS_ERR(dst)) return PTR_ERR(dst); skb_dst_set(skb, dst); diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index da385ae997a3..cf4555581282 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1110,9 +1110,14 @@ static int tcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg, flags = (msg->msg_flags & MSG_DONTWAIT) ? O_NONBLOCK : 0; err = __inet_stream_connect(sk->sk_socket, msg->msg_name, msg->msg_namelen, flags, 1); - inet->defer_connect = 0; - *copied = tp->fastopen_req->copied; - tcp_free_fastopen_req(tp); + /* fastopen_req could already be freed in __inet_stream_connect + * if the connection times out or gets rst + */ + if (tp->fastopen_req) { + *copied = tp->fastopen_req->copied; + tcp_free_fastopen_req(tp); + inet->defer_connect = 0; + } return err; } @@ -2318,6 +2323,10 @@ int tcp_disconnect(struct sock *sk, int flags) memset(&tp->rx_opt, 0, sizeof(tp->rx_opt)); __sk_dst_reset(sk); + /* Clean up fastopen related fields */ + tcp_free_fastopen_req(tp); + inet->defer_connect = 0; + WARN_ON(inet->inet_num && !icsk->icsk_bind_hash); sk->sk_error_report(sk); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 2c0ff327b6df..39c393cc0fd3 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5886,9 +5886,15 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb) if (th->syn) { if (th->fin) goto discard; - if (icsk->icsk_af_ops->conn_request(sk, skb) < 0) - return 1; + /* It is possible that we process SYN packets from backlog, + * so we need to make sure to disable BH right there. + */ + local_bh_disable(); + acceptable = icsk->icsk_af_ops->conn_request(sk, skb) >= 0; + local_bh_enable(); + if (!acceptable) + return 1; consume_skb(skb); return 0; } diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 77362b88a661..363172527e43 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -5693,13 +5693,18 @@ static int addrconf_sysctl_addr_gen_mode(struct ctl_table *ctl, int write, struct inet6_dev *idev = (struct inet6_dev *)ctl->extra1; struct net *net = (struct net *)ctl->extra2; + if (!rtnl_trylock()) + return restart_syscall(); + ret = proc_dointvec(ctl, write, buffer, lenp, ppos); if (write) { new_val = *((int *)ctl->data); - if (check_addr_gen_mode(new_val) < 0) - return -EINVAL; + if (check_addr_gen_mode(new_val) < 0) { + ret = -EINVAL; + goto out; + } /* request for default */ if (&net->ipv6.devconf_dflt->addr_gen_mode == ctl->data) { @@ -5708,20 +5713,23 @@ static int addrconf_sysctl_addr_gen_mode(struct ctl_table *ctl, int write, /* request for individual net device */ } else { if (!idev) - return ret; + goto out; - if (check_stable_privacy(idev, net, new_val) < 0) - return -EINVAL; + if (check_stable_privacy(idev, net, new_val) < 0) { + ret = -EINVAL; + goto out; + } if (idev->cnf.addr_gen_mode != new_val) { idev->cnf.addr_gen_mode = new_val; - rtnl_lock(); addrconf_dev_config(idev->dev); - rtnl_unlock(); } } } +out: + rtnl_unlock(); + return ret; } diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index 9948b5ce52da..986d4ca38832 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -589,6 +589,7 @@ int nf_ct_frag6_gather(struct net *net, struct sk_buff *skb, u32 user) hdr = ipv6_hdr(skb); fhdr = (struct frag_hdr *)skb_transport_header(skb); + skb_orphan(skb); fq = fq_find(net, fhdr->identification, user, &hdr->saddr, &hdr->daddr, skb->dev ? skb->dev->ifindex : 0, ip6_frag_ecn(hdr)); if (fq == NULL) { diff --git a/net/ipv6/route.c b/net/ipv6/route.c index f54f4265b37f..229bfcc451ef 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2169,10 +2169,13 @@ int ip6_del_rt(struct rt6_info *rt) static int __ip6_del_rt_siblings(struct rt6_info *rt, struct fib6_config *cfg) { struct nl_info *info = &cfg->fc_nlinfo; + struct net *net = info->nl_net; struct sk_buff *skb = NULL; struct fib6_table *table; - int err; + int err = -ENOENT; + if (rt == net->ipv6.ip6_null_entry) + goto out_put; table = rt->rt6i_table; write_lock_bh(&table->tb6_lock); @@ -2184,7 +2187,7 @@ static int __ip6_del_rt_siblings(struct rt6_info *rt, struct fib6_config *cfg) if (skb) { u32 seq = info->nlh ? info->nlh->nlmsg_seq : 0; - if (rt6_fill_node(info->nl_net, skb, rt, + if (rt6_fill_node(net, skb, rt, NULL, NULL, 0, RTM_DELROUTE, info->portid, seq, 0) < 0) { kfree_skb(skb); @@ -2198,17 +2201,18 @@ static int __ip6_del_rt_siblings(struct rt6_info *rt, struct fib6_config *cfg) rt6i_siblings) { err = fib6_del(sibling, info); if (err) - goto out; + goto out_unlock; } } err = fib6_del(rt, info); -out: +out_unlock: write_unlock_bh(&table->tb6_lock); +out_put: ip6_rt_put(rt); if (skb) { - rtnl_notify(skb, info->nl_net, info->portid, RTNLGRP_IPV6_ROUTE, + rtnl_notify(skb, net, info->portid, RTNLGRP_IPV6_ROUTE, info->nlh, gfp_any()); } return err; @@ -2891,6 +2895,7 @@ static const struct nla_policy rtm_ipv6_policy[RTA_MAX+1] = { [RTA_ENCAP] = { .type = NLA_NESTED }, [RTA_EXPIRES] = { .type = NLA_U32 }, [RTA_UID] = { .type = NLA_U32 }, + [RTA_MARK] = { .type = NLA_U32 }, }; static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh, @@ -3627,6 +3632,12 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh) rt = (struct rt6_info *)ip6_route_output(net, NULL, &fl6); } + if (rt == net->ipv6.ip6_null_entry) { + err = rt->dst.error; + ip6_rt_put(rt); + goto errout; + } + skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL); if (!skb) { ip6_rt_put(rt); diff --git a/net/mac80211/agg-rx.c b/net/mac80211/agg-rx.c index 3b5fd4188f2a..4456559cb056 100644 --- a/net/mac80211/agg-rx.c +++ b/net/mac80211/agg-rx.c @@ -85,7 +85,7 @@ void ___ieee80211_stop_rx_ba_session(struct sta_info *sta, u16 tid, ht_dbg(sta->sdata, "Rx BA session stop requested for %pM tid %u %s reason: %d\n", sta->sta.addr, tid, - initiator == WLAN_BACK_RECIPIENT ? "recipient" : "inititator", + initiator == WLAN_BACK_RECIPIENT ? "recipient" : "initiator", (int)reason); if (drv_ampdu_action(local, sta->sdata, ¶ms)) @@ -398,6 +398,7 @@ void __ieee80211_start_rx_ba_session(struct sta_info *sta, tid_agg_rx->timeout = timeout; tid_agg_rx->stored_mpdu_num = 0; tid_agg_rx->auto_seq = auto_seq; + tid_agg_rx->started = false; tid_agg_rx->reorder_buf_filtered = 0; status = WLAN_STATUS_SUCCESS; diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 159a1a733725..0e718437d080 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -428,7 +428,7 @@ struct ieee80211_sta_tx_tspec { bool downgraded; }; -DECLARE_EWMA(beacon_signal, 16, 4) +DECLARE_EWMA(beacon_signal, 4, 4) struct ieee80211_if_managed { struct timer_list timer; diff --git a/net/mac80211/pm.c b/net/mac80211/pm.c index 28a3a0957c9e..76a8bcd8ef11 100644 --- a/net/mac80211/pm.c +++ b/net/mac80211/pm.c @@ -168,6 +168,7 @@ int __ieee80211_suspend(struct ieee80211_hw *hw, struct cfg80211_wowlan *wowlan) break; } + flush_delayed_work(&sdata->dec_tailroom_needed_wk); drv_remove_interface(local, sdata); } diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 50ca3828b124..e48724a6725e 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -4,7 +4,7 @@ * Copyright 2006-2007 Jiri Benc <jbenc@suse.cz> * Copyright 2007-2010 Johannes Berg <johannes@sipsolutions.net> * Copyright 2013-2014 Intel Mobile Communications GmbH - * Copyright(c) 2015 - 2016 Intel Deutschland GmbH + * Copyright(c) 2015 - 2017 Intel Deutschland GmbH * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -1034,6 +1034,18 @@ static bool ieee80211_sta_manage_reorder_buf(struct ieee80211_sub_if_data *sdata buf_size = tid_agg_rx->buf_size; head_seq_num = tid_agg_rx->head_seq_num; + /* + * If the current MPDU's SN is smaller than the SSN, it shouldn't + * be reordered. + */ + if (unlikely(!tid_agg_rx->started)) { + if (ieee80211_sn_less(mpdu_seq_num, head_seq_num)) { + ret = false; + goto out; + } + tid_agg_rx->started = true; + } + /* frame with out of date sequence number */ if (ieee80211_sn_less(mpdu_seq_num, head_seq_num)) { dev_kfree_skb(skb); @@ -3880,6 +3892,7 @@ static bool ieee80211_invoke_fast_rx(struct ieee80211_rx_data *rx, stats->last_rate = sta_stats_encode_rate(status); stats->fragments++; + stats->packets++; if (!(status->flag & RX_FLAG_NO_SIGNAL_VAL)) { stats->last_signal = status->signal; @@ -4073,15 +4086,17 @@ static void __ieee80211_rx_handle_packet(struct ieee80211_hw *hw, ieee80211_is_beacon(hdr->frame_control))) ieee80211_scan_rx(local, skb); - if (pubsta) { - rx.sta = container_of(pubsta, struct sta_info, sta); - rx.sdata = rx.sta->sdata; - if (ieee80211_prepare_and_rx_handle(&rx, skb, true)) - return; - goto out; - } else if (ieee80211_is_data(fc)) { + if (ieee80211_is_data(fc)) { struct sta_info *sta, *prev_sta; + if (pubsta) { + rx.sta = container_of(pubsta, struct sta_info, sta); + rx.sdata = rx.sta->sdata; + if (ieee80211_prepare_and_rx_handle(&rx, skb, true)) + return; + goto out; + } + prev_sta = NULL; for_each_sta_info(local, hdr->addr2, sta, tmp) { diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index 4774e663a411..3323a2fb289b 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -688,7 +688,7 @@ static void __sta_info_recalc_tim(struct sta_info *sta, bool ignore_pending) } /* No need to do anything if the driver does all */ - if (ieee80211_hw_check(&local->hw, AP_LINK_PS)) + if (ieee80211_hw_check(&local->hw, AP_LINK_PS) && !local->ops->set_tim) return; if (sta->dead) @@ -1264,7 +1264,7 @@ void ieee80211_sta_ps_deliver_wakeup(struct sta_info *sta) sta_info_recalc_tim(sta); ps_dbg(sdata, - "STA %pM aid %d sending %d filtered/%d PS frames since STA not sleeping anymore\n", + "STA %pM aid %d sending %d filtered/%d PS frames since STA woke up\n", sta->sta.addr, sta->sta.aid, filtered, buffered); ieee80211_check_fast_xmit(sta); diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h index dd06ef0b8861..e65cda34d2bc 100644 --- a/net/mac80211/sta_info.h +++ b/net/mac80211/sta_info.h @@ -189,6 +189,7 @@ struct tid_ampdu_tx { * @auto_seq: used for offloaded BA sessions to automatically pick head_seq_and * and ssn. * @removed: this session is removed (but might have been found due to RCU) + * @started: this session has started (head ssn or higher was received) * * This structure's lifetime is managed by RCU, assignments to * the array holding it must hold the aggregation mutex. @@ -212,8 +213,9 @@ struct tid_ampdu_rx { u16 ssn; u16 buf_size; u16 timeout; - bool auto_seq; - bool removed; + u8 auto_seq:1, + removed:1, + started:1; }; /** @@ -370,7 +372,7 @@ struct mesh_sta { unsigned int fail_avg; }; -DECLARE_EWMA(signal, 1024, 8) +DECLARE_EWMA(signal, 10, 8) struct ieee80211_sta_rx_stats { unsigned long packets; diff --git a/net/mac80211/status.c b/net/mac80211/status.c index 0dd7c351002d..83b8b11f24ea 100644 --- a/net/mac80211/status.c +++ b/net/mac80211/status.c @@ -51,7 +51,8 @@ static void ieee80211_handle_filtered_frame(struct ieee80211_local *local, struct ieee80211_hdr *hdr = (void *)skb->data; int ac; - if (info->flags & IEEE80211_TX_CTL_NO_PS_BUFFER) { + if (info->flags & (IEEE80211_TX_CTL_NO_PS_BUFFER | + IEEE80211_TX_CTL_AMPDU)) { ieee80211_free_txskb(&local->hw, skb); return; } diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 24174c520239..0d17894798b5 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -1628,8 +1628,6 @@ static int __init nf_conntrack_sip_init(void) ports[ports_c++] = SIP_PORT; for (i = 0; i < ports_c; i++) { - memset(&sip[i], 0, sizeof(sip[i])); - nf_ct_helper_init(&sip[4 * i], AF_INET, IPPROTO_UDP, "sip", SIP_PORT, ports[i], i, sip_exp_policy, SIP_EXPECT_MAX, diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index ff7304ae58ac..5e0ccfd5bb37 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -461,16 +461,15 @@ nla_put_failure: return -1; } -static int nf_tables_table_notify(const struct nft_ctx *ctx, int event) +static void nf_tables_table_notify(const struct nft_ctx *ctx, int event) { struct sk_buff *skb; int err; if (!ctx->report && !nfnetlink_has_listeners(ctx->net, NFNLGRP_NFTABLES)) - return 0; + return; - err = -ENOBUFS; skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); if (skb == NULL) goto err; @@ -482,14 +481,11 @@ static int nf_tables_table_notify(const struct nft_ctx *ctx, int event) goto err; } - err = nfnetlink_send(skb, ctx->net, ctx->portid, NFNLGRP_NFTABLES, - ctx->report, GFP_KERNEL); + nfnetlink_send(skb, ctx->net, ctx->portid, NFNLGRP_NFTABLES, + ctx->report, GFP_KERNEL); + return; err: - if (err < 0) { - nfnetlink_set_err(ctx->net, ctx->portid, NFNLGRP_NFTABLES, - err); - } - return err; + nfnetlink_set_err(ctx->net, ctx->portid, NFNLGRP_NFTABLES, -ENOBUFS); } static int nf_tables_dump_tables(struct sk_buff *skb, @@ -1050,16 +1046,15 @@ nla_put_failure: return -1; } -static int nf_tables_chain_notify(const struct nft_ctx *ctx, int event) +static void nf_tables_chain_notify(const struct nft_ctx *ctx, int event) { struct sk_buff *skb; int err; if (!ctx->report && !nfnetlink_has_listeners(ctx->net, NFNLGRP_NFTABLES)) - return 0; + return; - err = -ENOBUFS; skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); if (skb == NULL) goto err; @@ -1072,14 +1067,11 @@ static int nf_tables_chain_notify(const struct nft_ctx *ctx, int event) goto err; } - err = nfnetlink_send(skb, ctx->net, ctx->portid, NFNLGRP_NFTABLES, - ctx->report, GFP_KERNEL); + nfnetlink_send(skb, ctx->net, ctx->portid, NFNLGRP_NFTABLES, + ctx->report, GFP_KERNEL); + return; err: - if (err < 0) { - nfnetlink_set_err(ctx->net, ctx->portid, NFNLGRP_NFTABLES, - err); - } - return err; + nfnetlink_set_err(ctx->net, ctx->portid, NFNLGRP_NFTABLES, -ENOBUFS); } static int nf_tables_dump_chains(struct sk_buff *skb, @@ -1934,18 +1926,16 @@ nla_put_failure: return -1; } -static int nf_tables_rule_notify(const struct nft_ctx *ctx, - const struct nft_rule *rule, - int event) +static void nf_tables_rule_notify(const struct nft_ctx *ctx, + const struct nft_rule *rule, int event) { struct sk_buff *skb; int err; if (!ctx->report && !nfnetlink_has_listeners(ctx->net, NFNLGRP_NFTABLES)) - return 0; + return; - err = -ENOBUFS; skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); if (skb == NULL) goto err; @@ -1958,14 +1948,11 @@ static int nf_tables_rule_notify(const struct nft_ctx *ctx, goto err; } - err = nfnetlink_send(skb, ctx->net, ctx->portid, NFNLGRP_NFTABLES, - ctx->report, GFP_KERNEL); + nfnetlink_send(skb, ctx->net, ctx->portid, NFNLGRP_NFTABLES, + ctx->report, GFP_KERNEL); + return; err: - if (err < 0) { - nfnetlink_set_err(ctx->net, ctx->portid, NFNLGRP_NFTABLES, - err); - } - return err; + nfnetlink_set_err(ctx->net, ctx->portid, NFNLGRP_NFTABLES, -ENOBUFS); } struct nft_rule_dump_ctx { @@ -2696,9 +2683,9 @@ nla_put_failure: return -1; } -static int nf_tables_set_notify(const struct nft_ctx *ctx, - const struct nft_set *set, - int event, gfp_t gfp_flags) +static void nf_tables_set_notify(const struct nft_ctx *ctx, + const struct nft_set *set, int event, + gfp_t gfp_flags) { struct sk_buff *skb; u32 portid = ctx->portid; @@ -2706,9 +2693,8 @@ static int nf_tables_set_notify(const struct nft_ctx *ctx, if (!ctx->report && !nfnetlink_has_listeners(ctx->net, NFNLGRP_NFTABLES)) - return 0; + return; - err = -ENOBUFS; skb = nlmsg_new(NLMSG_GOODSIZE, gfp_flags); if (skb == NULL) goto err; @@ -2719,12 +2705,11 @@ static int nf_tables_set_notify(const struct nft_ctx *ctx, goto err; } - err = nfnetlink_send(skb, ctx->net, portid, NFNLGRP_NFTABLES, - ctx->report, gfp_flags); + nfnetlink_send(skb, ctx->net, portid, NFNLGRP_NFTABLES, ctx->report, + gfp_flags); + return; err: - if (err < 0) - nfnetlink_set_err(ctx->net, portid, NFNLGRP_NFTABLES, err); - return err; + nfnetlink_set_err(ctx->net, portid, NFNLGRP_NFTABLES, -ENOBUFS); } static int nf_tables_dump_sets(struct sk_buff *skb, struct netlink_callback *cb) @@ -3504,10 +3489,10 @@ nla_put_failure: return -1; } -static int nf_tables_setelem_notify(const struct nft_ctx *ctx, - const struct nft_set *set, - const struct nft_set_elem *elem, - int event, u16 flags) +static void nf_tables_setelem_notify(const struct nft_ctx *ctx, + const struct nft_set *set, + const struct nft_set_elem *elem, + int event, u16 flags) { struct net *net = ctx->net; u32 portid = ctx->portid; @@ -3515,9 +3500,8 @@ static int nf_tables_setelem_notify(const struct nft_ctx *ctx, int err; if (!ctx->report && !nfnetlink_has_listeners(net, NFNLGRP_NFTABLES)) - return 0; + return; - err = -ENOBUFS; skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); if (skb == NULL) goto err; @@ -3529,12 +3513,11 @@ static int nf_tables_setelem_notify(const struct nft_ctx *ctx, goto err; } - err = nfnetlink_send(skb, net, portid, NFNLGRP_NFTABLES, ctx->report, - GFP_KERNEL); + nfnetlink_send(skb, net, portid, NFNLGRP_NFTABLES, ctx->report, + GFP_KERNEL); + return; err: - if (err < 0) - nfnetlink_set_err(net, portid, NFNLGRP_NFTABLES, err); - return err; + nfnetlink_set_err(net, portid, NFNLGRP_NFTABLES, -ENOBUFS); } static struct nft_trans *nft_trans_elem_alloc(struct nft_ctx *ctx, @@ -4476,18 +4459,17 @@ static int nf_tables_delobj(struct net *net, struct sock *nlsk, return nft_delobj(&ctx, obj); } -int nft_obj_notify(struct net *net, struct nft_table *table, - struct nft_object *obj, u32 portid, u32 seq, int event, - int family, int report, gfp_t gfp) +void nft_obj_notify(struct net *net, struct nft_table *table, + struct nft_object *obj, u32 portid, u32 seq, int event, + int family, int report, gfp_t gfp) { struct sk_buff *skb; int err; if (!report && !nfnetlink_has_listeners(net, NFNLGRP_NFTABLES)) - return 0; + return; - err = -ENOBUFS; skb = nlmsg_new(NLMSG_GOODSIZE, gfp); if (skb == NULL) goto err; @@ -4499,21 +4481,18 @@ int nft_obj_notify(struct net *net, struct nft_table *table, goto err; } - err = nfnetlink_send(skb, net, portid, NFNLGRP_NFTABLES, report, gfp); + nfnetlink_send(skb, net, portid, NFNLGRP_NFTABLES, report, gfp); + return; err: - if (err < 0) { - nfnetlink_set_err(net, portid, NFNLGRP_NFTABLES, err); - } - return err; + nfnetlink_set_err(net, portid, NFNLGRP_NFTABLES, -ENOBUFS); } EXPORT_SYMBOL_GPL(nft_obj_notify); -static int nf_tables_obj_notify(const struct nft_ctx *ctx, - struct nft_object *obj, int event) +static void nf_tables_obj_notify(const struct nft_ctx *ctx, + struct nft_object *obj, int event) { - return nft_obj_notify(ctx->net, ctx->table, obj, ctx->portid, - ctx->seq, event, ctx->afi->family, ctx->report, - GFP_KERNEL); + nft_obj_notify(ctx->net, ctx->table, obj, ctx->portid, ctx->seq, event, + ctx->afi->family, ctx->report, GFP_KERNEL); } static int nf_tables_fill_gen_info(struct sk_buff *skb, struct net *net, @@ -4543,7 +4522,8 @@ nla_put_failure: return -EMSGSIZE; } -static int nf_tables_gen_notify(struct net *net, struct sk_buff *skb, int event) +static void nf_tables_gen_notify(struct net *net, struct sk_buff *skb, + int event) { struct nlmsghdr *nlh = nlmsg_hdr(skb); struct sk_buff *skb2; @@ -4551,9 +4531,8 @@ static int nf_tables_gen_notify(struct net *net, struct sk_buff *skb, int event) if (nlmsg_report(nlh) && !nfnetlink_has_listeners(net, NFNLGRP_NFTABLES)) - return 0; + return; - err = -ENOBUFS; skb2 = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); if (skb2 == NULL) goto err; @@ -4565,14 +4544,12 @@ static int nf_tables_gen_notify(struct net *net, struct sk_buff *skb, int event) goto err; } - err = nfnetlink_send(skb2, net, NETLINK_CB(skb).portid, - NFNLGRP_NFTABLES, nlmsg_report(nlh), GFP_KERNEL); + nfnetlink_send(skb2, net, NETLINK_CB(skb).portid, NFNLGRP_NFTABLES, + nlmsg_report(nlh), GFP_KERNEL); + return; err: - if (err < 0) { - nfnetlink_set_err(net, NETLINK_CB(skb).portid, NFNLGRP_NFTABLES, - err); - } - return err; + nfnetlink_set_err(net, NETLINK_CB(skb).portid, NFNLGRP_NFTABLES, + -ENOBUFS); } static int nf_tables_getgen(struct net *net, struct sock *nlsk, diff --git a/net/netfilter/nft_set_rbtree.c b/net/netfilter/nft_set_rbtree.c index 71e8fb886a73..78dfbf9588b3 100644 --- a/net/netfilter/nft_set_rbtree.c +++ b/net/netfilter/nft_set_rbtree.c @@ -60,11 +60,10 @@ static bool nft_rbtree_lookup(const struct net *net, const struct nft_set *set, d = memcmp(this, key, set->klen); if (d < 0) { parent = parent->rb_left; - /* In case of adjacent ranges, we always see the high - * part of the range in first place, before the low one. - * So don't update interval if the keys are equal. - */ - if (interval && nft_rbtree_equal(set, this, interval)) + if (interval && + nft_rbtree_equal(set, this, interval) && + nft_rbtree_interval_end(this) && + !nft_rbtree_interval_end(interval)) continue; interval = rbe; } else if (d > 0) diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index b1beb2b94ec7..c82301ce3fff 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -796,9 +796,8 @@ static void ovs_fragment(struct net *net, struct vport *vport, unsigned long orig_dst; struct rt6_info ovs_rt; - if (!v6ops) { + if (!v6ops) goto err; - } prepare_frag(vport, skb, orig_network_offset, ovs_key_mac_proto(key)); diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c index 85cd59526670..e0a87776a010 100644 --- a/net/openvswitch/conntrack.c +++ b/net/openvswitch/conntrack.c @@ -485,7 +485,6 @@ static int handle_fragments(struct net *net, struct sw_flow_key *key, } else if (key->eth.type == htons(ETH_P_IPV6)) { enum ip6_defrag_users user = IP6_DEFRAG_CONNTRACK_IN + zone; - skb_orphan(skb); memset(IP6CB(skb), 0, sizeof(struct inet6_skb_parm)); err = nf_ct_frag6_gather(net, skb, user); if (err) { diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 2bd0d1949312..a0dbe7ca8f72 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -3103,7 +3103,7 @@ static int packet_bind_spkt(struct socket *sock, struct sockaddr *uaddr, int addr_len) { struct sock *sk = sock->sk; - char name[15]; + char name[sizeof(uaddr->sa_data) + 1]; /* * Check legality @@ -3111,7 +3111,11 @@ static int packet_bind_spkt(struct socket *sock, struct sockaddr *uaddr, if (addr_len != sizeof(struct sockaddr)) return -EINVAL; - strlcpy(name, uaddr->sa_data, sizeof(name)); + /* uaddr->sa_data comes from the userspace, it's not guaranteed to be + * zero-terminated. + */ + memcpy(name, uaddr->sa_data, sizeof(uaddr->sa_data)); + name[sizeof(uaddr->sa_data)] = 0; return packet_do_bind(sk, name, 0, pkt_sk(sk)->num); } diff --git a/net/rds/ib.c b/net/rds/ib.c index 91fe46f1e4cc..7a64c8db81ab 100644 --- a/net/rds/ib.c +++ b/net/rds/ib.c @@ -45,8 +45,8 @@ #include "ib.h" #include "ib_mr.h" -unsigned int rds_ib_mr_1m_pool_size = RDS_MR_1M_POOL_SIZE; -unsigned int rds_ib_mr_8k_pool_size = RDS_MR_8K_POOL_SIZE; +static unsigned int rds_ib_mr_1m_pool_size = RDS_MR_1M_POOL_SIZE; +static unsigned int rds_ib_mr_8k_pool_size = RDS_MR_8K_POOL_SIZE; unsigned int rds_ib_retry_count = RDS_IB_DEFAULT_RETRY_COUNT; module_param(rds_ib_mr_1m_pool_size, int, 0444); @@ -438,16 +438,12 @@ int rds_ib_init(void) if (ret) goto out_sysctl; - ret = rds_trans_register(&rds_ib_transport); - if (ret) - goto out_recv; + rds_trans_register(&rds_ib_transport); rds_info_register_func(RDS_INFO_IB_CONNECTIONS, rds_ib_ic_info); goto out; -out_recv: - rds_ib_recv_exit(); out_sysctl: rds_ib_sysctl_exit(); out_ibreg: diff --git a/net/rds/ib_mr.h b/net/rds/ib_mr.h index 24c086db4511..5d6e98a79a5e 100644 --- a/net/rds/ib_mr.h +++ b/net/rds/ib_mr.h @@ -107,8 +107,6 @@ struct rds_ib_mr_pool { }; extern struct workqueue_struct *rds_ib_mr_wq; -extern unsigned int rds_ib_mr_1m_pool_size; -extern unsigned int rds_ib_mr_8k_pool_size; extern bool prefer_frmr; struct rds_ib_mr_pool *rds_ib_create_mr_pool(struct rds_ib_device *rds_dev, diff --git a/net/rds/rds.h b/net/rds/rds.h index 966d2ee1f107..39518ef7af4d 100644 --- a/net/rds/rds.h +++ b/net/rds/rds.h @@ -903,7 +903,7 @@ void rds_connect_path_complete(struct rds_conn_path *conn, int curr); void rds_connect_complete(struct rds_connection *conn); /* transport.c */ -int rds_trans_register(struct rds_transport *trans); +void rds_trans_register(struct rds_transport *trans); void rds_trans_unregister(struct rds_transport *trans); struct rds_transport *rds_trans_get_preferred(struct net *net, __be32 addr); void rds_trans_put(struct rds_transport *trans); diff --git a/net/rds/tcp.c b/net/rds/tcp.c index 5438f6725092..a973d3b4dff0 100644 --- a/net/rds/tcp.c +++ b/net/rds/tcp.c @@ -652,16 +652,12 @@ static int rds_tcp_init(void) if (ret) goto out_pernet; - ret = rds_trans_register(&rds_tcp_transport); - if (ret) - goto out_recv; + rds_trans_register(&rds_tcp_transport); rds_info_register_func(RDS_INFO_TCP_SOCKETS, rds_tcp_tc_info); goto out; -out_recv: - rds_tcp_recv_exit(); out_pernet: unregister_pernet_subsys(&rds_tcp_net_ops); out_notifier: diff --git a/net/rds/transport.c b/net/rds/transport.c index 2ffd3e30c643..0b188dd0a344 100644 --- a/net/rds/transport.c +++ b/net/rds/transport.c @@ -40,7 +40,7 @@ static struct rds_transport *transports[RDS_TRANS_COUNT]; static DECLARE_RWSEM(rds_trans_sem); -int rds_trans_register(struct rds_transport *trans) +void rds_trans_register(struct rds_transport *trans) { BUG_ON(strlen(trans->t_name) + 1 > TRANSNAMSIZ); @@ -55,8 +55,6 @@ int rds_trans_register(struct rds_transport *trans) } up_write(&rds_trans_sem); - - return 0; } EXPORT_SYMBOL_GPL(rds_trans_register); diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 199b46e93e64..7fb59c3f1542 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -290,10 +290,11 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, cp.exclusive = false; cp.service_id = srx->srx_service; call = rxrpc_new_client_call(rx, &cp, srx, user_call_ID, gfp); + /* The socket has been unlocked. */ if (!IS_ERR(call)) call->notify_rx = notify_rx; - release_sock(&rx->sk); + mutex_unlock(&call->user_mutex); _leave(" = %p", call); return call; } @@ -310,7 +311,10 @@ EXPORT_SYMBOL(rxrpc_kernel_begin_call); void rxrpc_kernel_end_call(struct socket *sock, struct rxrpc_call *call) { _enter("%d{%d}", call->debug_id, atomic_read(&call->usage)); + + mutex_lock(&call->user_mutex); rxrpc_release_call(rxrpc_sk(sock->sk), call); + mutex_unlock(&call->user_mutex); rxrpc_put_call(call, rxrpc_call_put_kernel); } EXPORT_SYMBOL(rxrpc_kernel_end_call); @@ -450,14 +454,16 @@ static int rxrpc_sendmsg(struct socket *sock, struct msghdr *m, size_t len) case RXRPC_SERVER_BOUND: case RXRPC_SERVER_LISTENING: ret = rxrpc_do_sendmsg(rx, m, len); - break; + /* The socket has been unlocked */ + goto out; default: ret = -EINVAL; - break; + goto error_unlock; } error_unlock: release_sock(&rx->sk); +out: _leave(" = %d", ret); return ret; } diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 12be432be9b2..26a7b1db1361 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -467,6 +467,7 @@ struct rxrpc_call { struct rxrpc_connection *conn; /* connection carrying call */ struct rxrpc_peer *peer; /* Peer record for remote address */ struct rxrpc_sock __rcu *socket; /* socket responsible */ + struct mutex user_mutex; /* User access mutex */ ktime_t ack_at; /* When deferred ACK needs to happen */ ktime_t resend_at; /* When next resend needs to happen */ ktime_t ping_at; /* When next to send a ping */ diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c index 7c4c64ab8da2..0ed181f53f32 100644 --- a/net/rxrpc/call_accept.c +++ b/net/rxrpc/call_accept.c @@ -323,6 +323,8 @@ static struct rxrpc_call *rxrpc_alloc_incoming_call(struct rxrpc_sock *rx, * * If we want to report an error, we mark the skb with the packet type and * abort code and return NULL. + * + * The call is returned with the user access mutex held. */ struct rxrpc_call *rxrpc_new_incoming_call(struct rxrpc_local *local, struct rxrpc_connection *conn, @@ -371,6 +373,18 @@ found_service: trace_rxrpc_receive(call, rxrpc_receive_incoming, sp->hdr.serial, sp->hdr.seq); + /* Lock the call to prevent rxrpc_kernel_send/recv_data() and + * sendmsg()/recvmsg() inconveniently stealing the mutex once the + * notification is generated. + * + * The BUG should never happen because the kernel should be well + * behaved enough not to access the call before the first notification + * event and userspace is prevented from doing so until the state is + * appropriate. + */ + if (!mutex_trylock(&call->user_mutex)) + BUG(); + /* Make the call live. */ rxrpc_incoming_call(rx, call, skb); conn = call->conn; @@ -429,10 +443,12 @@ out: /* * handle acceptance of a call by userspace * - assign the user call ID to the call at the front of the queue + * - called with the socket locked. */ struct rxrpc_call *rxrpc_accept_call(struct rxrpc_sock *rx, unsigned long user_call_ID, rxrpc_notify_rx_t notify_rx) + __releases(&rx->sk.sk_lock.slock) { struct rxrpc_call *call; struct rb_node *parent, **pp; @@ -446,6 +462,7 @@ struct rxrpc_call *rxrpc_accept_call(struct rxrpc_sock *rx, if (list_empty(&rx->to_be_accepted)) { write_unlock(&rx->call_lock); + release_sock(&rx->sk); kleave(" = -ENODATA [empty]"); return ERR_PTR(-ENODATA); } @@ -470,10 +487,39 @@ struct rxrpc_call *rxrpc_accept_call(struct rxrpc_sock *rx, */ call = list_entry(rx->to_be_accepted.next, struct rxrpc_call, accept_link); + write_unlock(&rx->call_lock); + + /* We need to gain the mutex from the interrupt handler without + * upsetting lockdep, so we have to release it there and take it here. + * We are, however, still holding the socket lock, so other accepts + * must wait for us and no one can add the user ID behind our backs. + */ + if (mutex_lock_interruptible(&call->user_mutex) < 0) { + release_sock(&rx->sk); + kleave(" = -ERESTARTSYS"); + return ERR_PTR(-ERESTARTSYS); + } + + write_lock(&rx->call_lock); list_del_init(&call->accept_link); sk_acceptq_removed(&rx->sk); rxrpc_see_call(call); + /* Find the user ID insertion point. */ + pp = &rx->calls.rb_node; + parent = NULL; + while (*pp) { + parent = *pp; + call = rb_entry(parent, struct rxrpc_call, sock_node); + + if (user_call_ID < call->user_call_ID) + pp = &(*pp)->rb_left; + else if (user_call_ID > call->user_call_ID) + pp = &(*pp)->rb_right; + else + BUG(); + } + write_lock_bh(&call->state_lock); switch (call->state) { case RXRPC_CALL_SERVER_ACCEPTING: @@ -499,6 +545,7 @@ struct rxrpc_call *rxrpc_accept_call(struct rxrpc_sock *rx, write_unlock(&rx->call_lock); rxrpc_notify_socket(call); rxrpc_service_prealloc(rx, GFP_KERNEL); + release_sock(&rx->sk); _leave(" = %p{%d}", call, call->debug_id); return call; @@ -515,6 +562,7 @@ id_in_use: write_unlock(&rx->call_lock); out: rxrpc_service_prealloc(rx, GFP_KERNEL); + release_sock(&rx->sk); _leave(" = %d", ret); return ERR_PTR(ret); } diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c index 8b94db3c9b2e..d79cd36987a9 100644 --- a/net/rxrpc/call_object.c +++ b/net/rxrpc/call_object.c @@ -115,6 +115,7 @@ struct rxrpc_call *rxrpc_alloc_call(gfp_t gfp) if (!call->rxtx_annotations) goto nomem_2; + mutex_init(&call->user_mutex); setup_timer(&call->timer, rxrpc_call_timer_expired, (unsigned long)call); INIT_WORK(&call->processor, &rxrpc_process_call); @@ -194,14 +195,16 @@ static void rxrpc_start_call_timer(struct rxrpc_call *call) } /* - * set up a call for the given data - * - called in process context with IRQs enabled + * Set up a call for the given parameters. + * - Called with the socket lock held, which it must release. + * - If it returns a call, the call's lock will need releasing by the caller. */ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, struct rxrpc_conn_parameters *cp, struct sockaddr_rxrpc *srx, unsigned long user_call_ID, gfp_t gfp) + __releases(&rx->sk.sk_lock.slock) { struct rxrpc_call *call, *xcall; struct rb_node *parent, **pp; @@ -212,6 +215,7 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, call = rxrpc_alloc_client_call(srx, gfp); if (IS_ERR(call)) { + release_sock(&rx->sk); _leave(" = %ld", PTR_ERR(call)); return call; } @@ -219,6 +223,11 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, trace_rxrpc_call(call, rxrpc_call_new_client, atomic_read(&call->usage), here, (const void *)user_call_ID); + /* We need to protect a partially set up call against the user as we + * will be acting outside the socket lock. + */ + mutex_lock(&call->user_mutex); + /* Publish the call, even though it is incompletely set up as yet */ write_lock(&rx->call_lock); @@ -250,6 +259,9 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, list_add_tail(&call->link, &rxrpc_calls); write_unlock(&rxrpc_call_lock); + /* From this point on, the call is protected by its own lock. */ + release_sock(&rx->sk); + /* Set up or get a connection record and set the protocol parameters, * including channel number and call ID. */ @@ -279,6 +291,7 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, */ error_dup_user_ID: write_unlock(&rx->call_lock); + release_sock(&rx->sk); ret = -EEXIST; error: @@ -287,6 +300,7 @@ error: trace_rxrpc_call(call, rxrpc_call_error, atomic_read(&call->usage), here, ERR_PTR(ret)); rxrpc_release_call(rx, call); + mutex_unlock(&call->user_mutex); rxrpc_put_call(call, rxrpc_call_put); _leave(" = %d", ret); return ERR_PTR(ret); diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c index 78ec33477adf..9f4cfa25af7c 100644 --- a/net/rxrpc/input.c +++ b/net/rxrpc/input.c @@ -1194,6 +1194,7 @@ void rxrpc_data_ready(struct sock *udp_sk) goto reject_packet; } rxrpc_send_ping(call, skb, skew); + mutex_unlock(&call->user_mutex); } rxrpc_input_call_packet(call, skb, skew); diff --git a/net/rxrpc/recvmsg.c b/net/rxrpc/recvmsg.c index 28274a3c9831..6491ca46a03f 100644 --- a/net/rxrpc/recvmsg.c +++ b/net/rxrpc/recvmsg.c @@ -489,6 +489,20 @@ try_again: trace_rxrpc_recvmsg(call, rxrpc_recvmsg_dequeue, 0, 0, 0, 0); + /* We're going to drop the socket lock, so we need to lock the call + * against interference by sendmsg. + */ + if (!mutex_trylock(&call->user_mutex)) { + ret = -EWOULDBLOCK; + if (flags & MSG_DONTWAIT) + goto error_requeue_call; + ret = -ERESTARTSYS; + if (mutex_lock_interruptible(&call->user_mutex) < 0) + goto error_requeue_call; + } + + release_sock(&rx->sk); + if (test_bit(RXRPC_CALL_RELEASED, &call->flags)) BUG(); @@ -504,7 +518,7 @@ try_again: &call->user_call_ID); } if (ret < 0) - goto error; + goto error_unlock_call; } if (msg->msg_name) { @@ -535,12 +549,12 @@ try_again: } if (ret < 0) - goto error; + goto error_unlock_call; if (call->state == RXRPC_CALL_COMPLETE) { ret = rxrpc_recvmsg_term(call, msg); if (ret < 0) - goto error; + goto error_unlock_call; if (!(flags & MSG_PEEK)) rxrpc_release_call(rx, call); msg->msg_flags |= MSG_EOR; @@ -553,8 +567,21 @@ try_again: msg->msg_flags &= ~MSG_MORE; ret = copied; -error: +error_unlock_call: + mutex_unlock(&call->user_mutex); rxrpc_put_call(call, rxrpc_call_put); + trace_rxrpc_recvmsg(call, rxrpc_recvmsg_return, 0, 0, 0, ret); + return ret; + +error_requeue_call: + if (!(flags & MSG_PEEK)) { + write_lock_bh(&rx->recvmsg_lock); + list_add(&call->recvmsg_link, &rx->recvmsg_q); + write_unlock_bh(&rx->recvmsg_lock); + trace_rxrpc_recvmsg(call, rxrpc_recvmsg_requeue, 0, 0, 0, 0); + } else { + rxrpc_put_call(call, rxrpc_call_put); + } error_no_call: release_sock(&rx->sk); trace_rxrpc_recvmsg(call, rxrpc_recvmsg_return, 0, 0, 0, ret); @@ -611,7 +638,7 @@ int rxrpc_kernel_recv_data(struct socket *sock, struct rxrpc_call *call, iov.iov_len = size - *_offset; iov_iter_kvec(&iter, ITER_KVEC | READ, &iov, 1, size - *_offset); - lock_sock(sock->sk); + mutex_lock(&call->user_mutex); switch (call->state) { case RXRPC_CALL_CLIENT_RECV_REPLY: @@ -650,7 +677,7 @@ int rxrpc_kernel_recv_data(struct socket *sock, struct rxrpc_call *call, read_phase_complete: ret = 1; out: - release_sock(sock->sk); + mutex_unlock(&call->user_mutex); _leave(" = %d [%zu,%d]", ret, *_offset, *_abort); return ret; diff --git a/net/rxrpc/sendmsg.c b/net/rxrpc/sendmsg.c index 19b36c60fb4c..bc2d3dcff9de 100644 --- a/net/rxrpc/sendmsg.c +++ b/net/rxrpc/sendmsg.c @@ -61,9 +61,12 @@ static int rxrpc_wait_for_tx_window(struct rxrpc_sock *rx, } trace_rxrpc_transmit(call, rxrpc_transmit_wait); - release_sock(&rx->sk); + mutex_unlock(&call->user_mutex); *timeo = schedule_timeout(*timeo); - lock_sock(&rx->sk); + if (mutex_lock_interruptible(&call->user_mutex) < 0) { + ret = sock_intr_errno(*timeo); + break; + } } remove_wait_queue(&call->waitq, &myself); @@ -173,7 +176,7 @@ static void rxrpc_queue_packet(struct rxrpc_call *call, struct sk_buff *skb, /* * send data through a socket * - must be called in process context - * - caller holds the socket locked + * - The caller holds the call user access mutex, but not the socket lock. */ static int rxrpc_send_data(struct rxrpc_sock *rx, struct rxrpc_call *call, @@ -439,10 +442,13 @@ static int rxrpc_sendmsg_cmsg(struct msghdr *msg, /* * Create a new client call for sendmsg(). + * - Called with the socket lock held, which it must release. + * - If it returns a call, the call's lock will need releasing by the caller. */ static struct rxrpc_call * rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, unsigned long user_call_ID, bool exclusive) + __releases(&rx->sk.sk_lock.slock) { struct rxrpc_conn_parameters cp; struct rxrpc_call *call; @@ -452,8 +458,10 @@ rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, _enter(""); - if (!msg->msg_name) + if (!msg->msg_name) { + release_sock(&rx->sk); return ERR_PTR(-EDESTADDRREQ); + } key = rx->key; if (key && !rx->key->payload.data[0]) @@ -466,6 +474,7 @@ rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, cp.exclusive = rx->exclusive | exclusive; cp.service_id = srx->srx_service; call = rxrpc_new_client_call(rx, &cp, srx, user_call_ID, GFP_KERNEL); + /* The socket is now unlocked */ _leave(" = %p\n", call); return call; @@ -477,6 +486,7 @@ rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, * - the socket may be either a client socket or a server socket */ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) + __releases(&rx->sk.sk_lock.slock) { enum rxrpc_command cmd; struct rxrpc_call *call; @@ -490,12 +500,14 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) ret = rxrpc_sendmsg_cmsg(msg, &user_call_ID, &cmd, &abort_code, &exclusive); if (ret < 0) - return ret; + goto error_release_sock; if (cmd == RXRPC_CMD_ACCEPT) { + ret = -EINVAL; if (rx->sk.sk_state != RXRPC_SERVER_LISTENING) - return -EINVAL; + goto error_release_sock; call = rxrpc_accept_call(rx, user_call_ID, NULL); + /* The socket is now unlocked. */ if (IS_ERR(call)) return PTR_ERR(call); rxrpc_put_call(call, rxrpc_call_put); @@ -504,12 +516,30 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) call = rxrpc_find_call_by_user_ID(rx, user_call_ID); if (!call) { + ret = -EBADSLT; if (cmd != RXRPC_CMD_SEND_DATA) - return -EBADSLT; + goto error_release_sock; call = rxrpc_new_client_call_for_sendmsg(rx, msg, user_call_ID, exclusive); + /* The socket is now unlocked... */ if (IS_ERR(call)) return PTR_ERR(call); + /* ... and we have the call lock. */ + } else { + ret = -EBUSY; + if (call->state == RXRPC_CALL_UNINITIALISED || + call->state == RXRPC_CALL_CLIENT_AWAIT_CONN || + call->state == RXRPC_CALL_SERVER_PREALLOC || + call->state == RXRPC_CALL_SERVER_SECURING || + call->state == RXRPC_CALL_SERVER_ACCEPTING) + goto error_release_sock; + + ret = mutex_lock_interruptible(&call->user_mutex); + release_sock(&rx->sk); + if (ret < 0) { + ret = -ERESTARTSYS; + goto error_put; + } } _debug("CALL %d USR %lx ST %d on CONN %p", @@ -537,9 +567,15 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) ret = rxrpc_send_data(rx, call, msg, len); } + mutex_unlock(&call->user_mutex); +error_put: rxrpc_put_call(call, rxrpc_call_put); _leave(" = %d", ret); return ret; + +error_release_sock: + release_sock(&rx->sk); + return ret; } /** @@ -564,7 +600,7 @@ int rxrpc_kernel_send_data(struct socket *sock, struct rxrpc_call *call, ASSERTCMP(msg->msg_name, ==, NULL); ASSERTCMP(msg->msg_control, ==, NULL); - lock_sock(sock->sk); + mutex_lock(&call->user_mutex); _debug("CALL %d USR %lx ST %d on CONN %p", call->debug_id, call->user_call_ID, call->state, call->conn); @@ -579,7 +615,7 @@ int rxrpc_kernel_send_data(struct socket *sock, struct rxrpc_call *call, ret = rxrpc_send_data(rxrpc_sk(sock->sk), call, msg, len); } - release_sock(sock->sk); + mutex_unlock(&call->user_mutex); _leave(" = %d", ret); return ret; } @@ -600,12 +636,12 @@ void rxrpc_kernel_abort_call(struct socket *sock, struct rxrpc_call *call, { _enter("{%d},%d,%d,%s", call->debug_id, abort_code, error, why); - lock_sock(sock->sk); + mutex_lock(&call->user_mutex); if (rxrpc_abort_call(why, call, 0, abort_code, error)) rxrpc_send_abort_packet(call); - release_sock(sock->sk); + mutex_unlock(&call->user_mutex); _leave(""); } diff --git a/net/sctp/input.c b/net/sctp/input.c index fc458968fe4b..2a28ab20487f 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -884,14 +884,17 @@ int sctp_hash_transport(struct sctp_transport *t) arg.paddr = &t->ipaddr; arg.lport = htons(t->asoc->base.bind_addr.port); + rcu_read_lock(); list = rhltable_lookup(&sctp_transport_hashtable, &arg, sctp_hash_params); rhl_for_each_entry_rcu(transport, tmp, list, node) if (transport->asoc->ep == t->asoc->ep) { + rcu_read_unlock(); err = -EEXIST; goto out; } + rcu_read_unlock(); err = rhltable_insert_key(&sctp_transport_hashtable, &arg, &t->node, sctp_hash_params); diff --git a/net/strparser/strparser.c b/net/strparser/strparser.c index 41adf362936d..b5c279b22680 100644 --- a/net/strparser/strparser.c +++ b/net/strparser/strparser.c @@ -504,6 +504,7 @@ static int __init strp_mod_init(void) static void __exit strp_mod_exit(void) { + destroy_workqueue(strp_wq); } module_init(strp_mod_init); module_exit(strp_mod_exit); |