From e70a3aad44cc8b24986687ffc98c4a4f6ecf25ea Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Thu, 2 Aug 2018 23:20:38 -0700 Subject: ipv6: fix double refcount of fib6_metrics All the callers of ip6_rt_copy_init()/rt6_set_from() hold refcnt of the "from" fib6_info, so there is no need to hold fib6_metrics refcnt again, because fib6_metrics refcnt is only released when fib6_info is gone, that is, they have the same life time, so the whole fib6_metrics refcnt can be removed actually. This fixes a kmemleak warning reported by Sabrina. Fixes: 93531c674315 ("net/ipv6: separate handling of FIB entries from dst based routes") Reported-by: Sabrina Dubroca Cc: Sabrina Dubroca Cc: David Ahern Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- net/ipv6/route.c | 4 ---- 1 file changed, 4 deletions(-) (limited to 'net') diff --git a/net/ipv6/route.c b/net/ipv6/route.c index ec18b3ce8b6d..7208c16302f6 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -978,10 +978,6 @@ static void rt6_set_from(struct rt6_info *rt, struct fib6_info *from) rt->rt6i_flags &= ~RTF_EXPIRES; rcu_assign_pointer(rt->from, from); dst_init_metrics(&rt->dst, from->fib6_metrics->metrics, true); - if (from->fib6_metrics != &dst_default_metrics) { - rt->dst._metrics |= DST_METRICS_REFCOUNTED; - refcount_inc(&from->fib6_metrics->refcnt); - } } /* Caller must already hold reference to @ort */ -- cgit v1.2.3 From 82a40777de12728dedf4075453b694f0d1baee80 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Sun, 5 Aug 2018 22:46:07 +0800 Subject: ip6_tunnel: use the right value for ipv4 min mtu check in ip6_tnl_xmit According to RFC791, 68 bytes is the minimum size of IPv4 datagram every device must be able to forward without further fragmentation while 576 bytes is the minimum size of IPv4 datagram every device has to be able to receive, so in ip6_tnl_xmit(), 68(IPV4_MIN_MTU) should be the right value for the ipv4 min mtu check in ip6_tnl_xmit. While at it, change to use max() instead of if statement. Fixes: c9fefa08190f ("ip6_tunnel: get the min mtu properly in ip6_tnl_xmit") Reported-by: Sabrina Dubroca Signed-off-by: Xin Long Signed-off-by: David S. Miller --- net/ipv6/ip6_tunnel.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) (limited to 'net') diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 00e138a44cbb..1cc9650af9fb 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -1133,12 +1133,8 @@ route_lookup: max_headroom += 8; mtu -= 8; } - if (skb->protocol == htons(ETH_P_IPV6)) { - if (mtu < IPV6_MIN_MTU) - mtu = IPV6_MIN_MTU; - } else if (mtu < 576) { - mtu = 576; - } + mtu = max(mtu, skb->protocol == htons(ETH_P_IPV6) ? + IPV6_MIN_MTU : IPV4_MIN_MTU); skb_dst_update_pmtu(skb, mtu); if (skb->len - t->tun_hlen - eth_hlen > mtu && !skb_is_gso(skb)) { -- cgit v1.2.3 From 4576cd469d980317c4edd9173f8b694aa71ea3a3 Mon Sep 17 00:00:00 2001 From: Willem de Bruijn Date: Mon, 6 Aug 2018 10:38:34 -0400 Subject: packet: refine ring v3 block size test to hold one frame TPACKET_V3 stores variable length frames in fixed length blocks. Blocks must be able to store a block header, optional private space and at least one minimum sized frame. Frames, even for a zero snaplen packet, store metadata headers and optional reserved space. In the block size bounds check, ensure that the frame of the chosen configuration fits. This includes sockaddr_ll and optional tp_reserve. Syzbot was able to construct a ring with insuffient room for the sockaddr_ll in the header of a zero-length frame, triggering an out-of-bounds write in dev_parse_header. Convert the comparison to less than, as zero is a valid snap len. This matches the test for minimum tp_frame_size immediately below. Fixes: f6fb8f100b80 ("af-packet: TPACKET_V3 flexible buffer implementation.") Fixes: eb73190f4fbe ("net/packet: refine check for priv area size") Reported-by: syzbot Signed-off-by: Willem de Bruijn Signed-off-by: David S. Miller --- net/packet/af_packet.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'net') diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 9b27d0cd766d..e6445d8f3f57 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -4226,6 +4226,8 @@ static int packet_set_ring(struct sock *sk, union tpacket_req_u *req_u, } if (req->tp_block_nr) { + unsigned int min_frame_size; + /* Sanity tests and some calculations */ err = -EBUSY; if (unlikely(rb->pg_vec)) @@ -4248,12 +4250,12 @@ static int packet_set_ring(struct sock *sk, union tpacket_req_u *req_u, goto out; if (unlikely(!PAGE_ALIGNED(req->tp_block_size))) goto out; + min_frame_size = po->tp_hdrlen + po->tp_reserve; if (po->tp_version >= TPACKET_V3 && - req->tp_block_size <= - BLK_PLUS_PRIV((u64)req_u->req3.tp_sizeof_priv) + sizeof(struct tpacket3_hdr)) + req->tp_block_size < + BLK_PLUS_PRIV((u64)req_u->req3.tp_sizeof_priv) + min_frame_size) goto out; - if (unlikely(req->tp_frame_size < po->tp_hdrlen + - po->tp_reserve)) + if (unlikely(req->tp_frame_size < min_frame_size)) goto out; if (unlikely(req->tp_frame_size & (TPACKET_ALIGNMENT - 1))) goto out; -- cgit v1.2.3 From 455f05ecd2b219e9a216050796d30c830d9bc393 Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Mon, 6 Aug 2018 11:06:02 -0700 Subject: vsock: split dwork to avoid reinitializations syzbot reported that we reinitialize an active delayed work in vsock_stream_connect(): ODEBUG: init active (active state 0) object type: timer_list hint: delayed_work_timer_fn+0x0/0x90 kernel/workqueue.c:1414 WARNING: CPU: 1 PID: 11518 at lib/debugobjects.c:329 debug_print_object+0x16a/0x210 lib/debugobjects.c:326 The pattern is apparently wrong, we should only initialize the dealyed work once and could repeatly schedule it. So we have to move out the initializations to allocation side. And to avoid confusion, we can split the shared dwork into two, instead of re-using the same one. Fixes: d021c344051a ("VSOCK: Introduce VM Sockets") Reported-by: Cc: Andy king Cc: Stefan Hajnoczi Cc: Jorgen Hansen Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- include/net/af_vsock.h | 4 ++-- net/vmw_vsock/af_vsock.c | 15 ++++++++------- net/vmw_vsock/vmci_transport.c | 3 +-- 3 files changed, 11 insertions(+), 11 deletions(-) (limited to 'net') diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h index 9324ac2d9ff2..43913ae79f64 100644 --- a/include/net/af_vsock.h +++ b/include/net/af_vsock.h @@ -64,7 +64,8 @@ struct vsock_sock { struct list_head pending_links; struct list_head accept_queue; bool rejected; - struct delayed_work dwork; + struct delayed_work connect_work; + struct delayed_work pending_work; struct delayed_work close_work; bool close_work_scheduled; u32 peer_shutdown; @@ -77,7 +78,6 @@ struct vsock_sock { s64 vsock_stream_has_data(struct vsock_sock *vsk); s64 vsock_stream_has_space(struct vsock_sock *vsk); -void vsock_pending_work(struct work_struct *work); struct sock *__vsock_create(struct net *net, struct socket *sock, struct sock *parent, diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index c1076c19b858..ab27a2872935 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -451,14 +451,14 @@ static int vsock_send_shutdown(struct sock *sk, int mode) return transport->shutdown(vsock_sk(sk), mode); } -void vsock_pending_work(struct work_struct *work) +static void vsock_pending_work(struct work_struct *work) { struct sock *sk; struct sock *listener; struct vsock_sock *vsk; bool cleanup; - vsk = container_of(work, struct vsock_sock, dwork.work); + vsk = container_of(work, struct vsock_sock, pending_work.work); sk = sk_vsock(vsk); listener = vsk->listener; cleanup = true; @@ -498,7 +498,6 @@ out: sock_put(sk); sock_put(listener); } -EXPORT_SYMBOL_GPL(vsock_pending_work); /**** SOCKET OPERATIONS ****/ @@ -597,6 +596,8 @@ static int __vsock_bind(struct sock *sk, struct sockaddr_vm *addr) return retval; } +static void vsock_connect_timeout(struct work_struct *work); + struct sock *__vsock_create(struct net *net, struct socket *sock, struct sock *parent, @@ -638,6 +639,8 @@ struct sock *__vsock_create(struct net *net, vsk->sent_request = false; vsk->ignore_connecting_rst = false; vsk->peer_shutdown = 0; + INIT_DELAYED_WORK(&vsk->connect_work, vsock_connect_timeout); + INIT_DELAYED_WORK(&vsk->pending_work, vsock_pending_work); psk = parent ? vsock_sk(parent) : NULL; if (parent) { @@ -1117,7 +1120,7 @@ static void vsock_connect_timeout(struct work_struct *work) struct vsock_sock *vsk; int cancel = 0; - vsk = container_of(work, struct vsock_sock, dwork.work); + vsk = container_of(work, struct vsock_sock, connect_work.work); sk = sk_vsock(vsk); lock_sock(sk); @@ -1221,9 +1224,7 @@ static int vsock_stream_connect(struct socket *sock, struct sockaddr *addr, * timeout fires. */ sock_hold(sk); - INIT_DELAYED_WORK(&vsk->dwork, - vsock_connect_timeout); - schedule_delayed_work(&vsk->dwork, timeout); + schedule_delayed_work(&vsk->connect_work, timeout); /* Skip ahead to preserve error code set above. */ goto out_wait; diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c index a7a73ffe675b..cb332adb84cd 100644 --- a/net/vmw_vsock/vmci_transport.c +++ b/net/vmw_vsock/vmci_transport.c @@ -1094,8 +1094,7 @@ static int vmci_transport_recv_listen(struct sock *sk, vpending->listener = sk; sock_hold(sk); sock_hold(pending); - INIT_DELAYED_WORK(&vpending->dwork, vsock_pending_work); - schedule_delayed_work(&vpending->dwork, HZ); + schedule_delayed_work(&vpending->pending_work, HZ); out: return err; -- cgit v1.2.3 From 37436d9c0e8f62c3eebe204ff5776ff31fd64658 Mon Sep 17 00:00:00 2001 From: Ying Xue Date: Tue, 7 Aug 2018 15:52:32 +0800 Subject: tipc: fix an interrupt unsafe locking scenario Commit 9faa89d4ed9d ("tipc: make function tipc_net_finalize() thread safe") tries to make it thread safe to set node address, so it uses node_list_lock lock to serialize the whole process of setting node address in tipc_net_finalize(). But it causes the following interrupt unsafe locking scenario: CPU0 CPU1 ---- ---- rht_deferred_worker() rhashtable_rehash_table() lock(&(&ht->lock)->rlock) tipc_nl_compat_doit() tipc_net_finalize() local_irq_disable(); lock(&(&tn->node_list_lock)->rlock); tipc_sk_reinit() rhashtable_walk_enter() lock(&(&ht->lock)->rlock); tipc_disc_rcv() tipc_node_check_dest() tipc_node_create() lock(&(&tn->node_list_lock)->rlock); *** DEADLOCK *** When rhashtable_rehash_table() holds ht->lock on CPU0, it doesn't disable BH. So if an interrupt happens after the lock, it can create an inverse lock ordering between ht->lock and tn->node_list_lock. As a consequence, deadlock might happen. The reason causing the inverse lock ordering scenario above is because the initial purpose of node_list_lock is not designed to do the serialization of node address setting. As cmpxchg() can guarantee CAS (compare-and-swap) process is atomic, we use it to replace node_list_lock to ensure setting node address can be atomically finished. It turns out the potential deadlock can be avoided as well. Fixes: 9faa89d4ed9d ("tipc: make function tipc_net_finalize() thread safe") Signed-off-by: Ying Xue Acked-by: Jon Maloy Signed-off-by: David S. Miller --- net/tipc/net.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'net') diff --git a/net/tipc/net.c b/net/tipc/net.c index a7f6964c3a4b..62199cf5a56c 100644 --- a/net/tipc/net.c +++ b/net/tipc/net.c @@ -123,15 +123,13 @@ void tipc_net_finalize(struct net *net, u32 addr) { struct tipc_net *tn = tipc_net(net); - spin_lock_bh(&tn->node_list_lock); - if (!tipc_own_addr(net)) { + if (!cmpxchg(&tn->node_addr, 0, addr)) { tipc_set_node_addr(net, addr); tipc_named_reinit(net); tipc_sk_reinit(net); tipc_nametbl_publish(net, TIPC_CFG_SRV, addr, addr, TIPC_CLUSTER_SCOPE, 0, addr); } - spin_unlock_bh(&tn->node_list_lock); } void tipc_net_stop(struct net *net) -- cgit v1.2.3 From 61ef4b07fcdc30535889990cf4229766502561cf Mon Sep 17 00:00:00 2001 From: Alexey Kodanev Date: Tue, 7 Aug 2018 20:03:57 +0300 Subject: dccp: fix undefined behavior with 'cwnd' shift in ccid2_cwnd_restart() The shift of 'cwnd' with '(now - hc->tx_lsndtime) / hc->tx_rto' value can lead to undefined behavior [1]. In order to fix this use a gradual shift of the window with a 'while' loop, similar to what tcp_cwnd_restart() is doing. When comparing delta and RTO there is a minor difference between TCP and DCCP, the last one also invokes dccp_cwnd_restart() and reduces 'cwnd' if delta equals RTO. That case is preserved in this change. [1]: [40850.963623] UBSAN: Undefined behaviour in net/dccp/ccids/ccid2.c:237:7 [40851.043858] shift exponent 67 is too large for 32-bit type 'unsigned int' [40851.127163] CPU: 3 PID: 15940 Comm: netstress Tainted: G W E 4.18.0-rc7.x86_64 #1 ... [40851.377176] Call Trace: [40851.408503] dump_stack+0xf1/0x17b [40851.451331] ? show_regs_print_info+0x5/0x5 [40851.503555] ubsan_epilogue+0x9/0x7c [40851.548363] __ubsan_handle_shift_out_of_bounds+0x25b/0x2b4 [40851.617109] ? __ubsan_handle_load_invalid_value+0x18f/0x18f [40851.686796] ? xfrm4_output_finish+0x80/0x80 [40851.739827] ? lock_downgrade+0x6d0/0x6d0 [40851.789744] ? xfrm4_prepare_output+0x160/0x160 [40851.845912] ? ip_queue_xmit+0x810/0x1db0 [40851.895845] ? ccid2_hc_tx_packet_sent+0xd36/0x10a0 [dccp] [40851.963530] ccid2_hc_tx_packet_sent+0xd36/0x10a0 [dccp] [40852.029063] dccp_xmit_packet+0x1d3/0x720 [dccp] [40852.086254] dccp_write_xmit+0x116/0x1d0 [dccp] [40852.142412] dccp_sendmsg+0x428/0xb20 [dccp] [40852.195454] ? inet_dccp_listen+0x200/0x200 [dccp] [40852.254833] ? sched_clock+0x5/0x10 [40852.298508] ? sched_clock+0x5/0x10 [40852.342194] ? inet_create+0xdf0/0xdf0 [40852.388988] sock_sendmsg+0xd9/0x160 ... Fixes: 113ced1f52e5 ("dccp ccid-2: Perform congestion-window validation") Signed-off-by: Alexey Kodanev Signed-off-by: David S. Miller --- net/dccp/ccids/ccid2.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/dccp/ccids/ccid2.c b/net/dccp/ccids/ccid2.c index 2b75df469220..842a9c7c73a3 100644 --- a/net/dccp/ccids/ccid2.c +++ b/net/dccp/ccids/ccid2.c @@ -229,14 +229,16 @@ static void ccid2_cwnd_restart(struct sock *sk, const u32 now) struct ccid2_hc_tx_sock *hc = ccid2_hc_tx_sk(sk); u32 cwnd = hc->tx_cwnd, restart_cwnd, iwnd = rfc3390_bytes_to_packets(dccp_sk(sk)->dccps_mss_cache); + s32 delta = now - hc->tx_lsndtime; hc->tx_ssthresh = max(hc->tx_ssthresh, (cwnd >> 1) + (cwnd >> 2)); /* don't reduce cwnd below the initial window (IW) */ restart_cwnd = min(cwnd, iwnd); - cwnd >>= (now - hc->tx_lsndtime) / hc->tx_rto; - hc->tx_cwnd = max(cwnd, restart_cwnd); + while ((delta -= hc->tx_rto) >= 0 && cwnd > restart_cwnd) + cwnd >>= 1; + hc->tx_cwnd = max(cwnd, restart_cwnd); hc->tx_cwnd_stamp = now; hc->tx_cwnd_used = 0; -- cgit v1.2.3 From 0dcb82254d65f72333aa50ad626d1e9665ad093b Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Tue, 7 Aug 2018 12:41:38 -0700 Subject: llc: use refcount_inc_not_zero() for llc_sap_find() llc_sap_put() decreases the refcnt before deleting sap from the global list. Therefore, there is a chance llc_sap_find() could find a sap with zero refcnt in this global list. Close this race condition by checking if refcnt is zero or not in llc_sap_find(), if it is zero then it is being removed so we can just treat it as gone. Reported-by: Signed-off-by: Cong Wang Signed-off-by: David S. Miller --- include/net/llc.h | 5 +++++ net/llc/llc_core.c | 4 ++-- 2 files changed, 7 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/include/net/llc.h b/include/net/llc.h index dc35f25eb679..890a87318014 100644 --- a/include/net/llc.h +++ b/include/net/llc.h @@ -116,6 +116,11 @@ static inline void llc_sap_hold(struct llc_sap *sap) refcount_inc(&sap->refcnt); } +static inline bool llc_sap_hold_safe(struct llc_sap *sap) +{ + return refcount_inc_not_zero(&sap->refcnt); +} + void llc_sap_close(struct llc_sap *sap); static inline void llc_sap_put(struct llc_sap *sap) diff --git a/net/llc/llc_core.c b/net/llc/llc_core.c index 89041260784c..260b3dc1b4a2 100644 --- a/net/llc/llc_core.c +++ b/net/llc/llc_core.c @@ -73,8 +73,8 @@ struct llc_sap *llc_sap_find(unsigned char sap_value) rcu_read_lock_bh(); sap = __llc_sap_find(sap_value); - if (sap) - llc_sap_hold(sap); + if (!sap || !llc_sap_hold_safe(sap)) + sap = NULL; rcu_read_unlock_bh(); return sap; } -- cgit v1.2.3 From 330bdcfadceea5e9a1526d731711e163f9a90975 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 8 Aug 2018 11:30:02 +0100 Subject: rxrpc: Fix the keepalive generator [ver #2] AF_RXRPC has a keepalive message generator that generates a message for a peer ~20s after the last transmission to that peer to keep firewall ports open. The implementation is incorrect in the following ways: (1) It mixes up ktime_t and time64_t types. (2) It uses ktime_get_real(), the output of which may jump forward or backward due to adjustments to the time of day. (3) If the current time jumps forward too much or jumps backwards, the generator function will crank the base of the time ring round one slot at a time (ie. a 1s period) until it catches up, spewing out VERSION packets as it goes. Fix the problem by: (1) Only using time64_t. There's no need for sub-second resolution. (2) Use ktime_get_seconds() rather than ktime_get_real() so that time isn't perceived to go backwards. (3) Simplifying rxrpc_peer_keepalive_worker() by splitting it into two parts: (a) The "worker" function that manages the buckets and the timer. (b) The "dispatch" function that takes the pending peers and potentially transmits a keepalive packet before putting them back in the ring into the slot appropriate to the revised last-Tx time. (4) Taking everything that's pending out of the ring and splicing it into a temporary collector list for processing. In the case that there's been a significant jump forward, the ring gets entirely emptied and then the time base can be warped forward before the peers are processed. The warping can't happen if the ring isn't empty because the slot a peer is in is keepalive-time dependent, relative to the base time. (5) Limit the number of iterations of the bucket array when scanning it. (6) Set the timer to skip any empty slots as there's no point waking up if there's nothing to do yet. This can be triggered by an incoming call from a server after a reboot with AF_RXRPC and AFS built into the kernel causing a peer record to be set up before userspace is started. The system clock is then adjusted by userspace, thereby potentially causing the keepalive generator to have a meltdown - which leads to a message like: watchdog: BUG: soft lockup - CPU#0 stuck for 23s! [kworker/0:1:23] ... Workqueue: krxrpcd rxrpc_peer_keepalive_worker EIP: lock_acquire+0x69/0x80 ... Call Trace: ? rxrpc_peer_keepalive_worker+0x5e/0x350 ? _raw_spin_lock_bh+0x29/0x60 ? rxrpc_peer_keepalive_worker+0x5e/0x350 ? rxrpc_peer_keepalive_worker+0x5e/0x350 ? __lock_acquire+0x3d3/0x870 ? process_one_work+0x110/0x340 ? process_one_work+0x166/0x340 ? process_one_work+0x110/0x340 ? worker_thread+0x39/0x3c0 ? kthread+0xdb/0x110 ? cancel_delayed_work+0x90/0x90 ? kthread_stop+0x70/0x70 ? ret_from_fork+0x19/0x24 Fixes: ace45bec6d77 ("rxrpc: Fix firewall route keepalive") Reported-by: kernel test robot Signed-off-by: David Howells Signed-off-by: David S. Miller --- net/rxrpc/ar-internal.h | 8 +-- net/rxrpc/conn_event.c | 4 +- net/rxrpc/net_ns.c | 6 +- net/rxrpc/output.c | 12 ++-- net/rxrpc/peer_event.c | 156 +++++++++++++++++++++++++++--------------------- net/rxrpc/peer_object.c | 8 +-- net/rxrpc/rxkad.c | 4 +- 7 files changed, 109 insertions(+), 89 deletions(-) (limited to 'net') diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 5fb7d3254d9e..707630ab4713 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -104,9 +104,9 @@ struct rxrpc_net { #define RXRPC_KEEPALIVE_TIME 20 /* NAT keepalive time in seconds */ u8 peer_keepalive_cursor; - ktime_t peer_keepalive_base; - struct hlist_head peer_keepalive[RXRPC_KEEPALIVE_TIME + 1]; - struct hlist_head peer_keepalive_new; + time64_t peer_keepalive_base; + struct list_head peer_keepalive[32]; + struct list_head peer_keepalive_new; struct timer_list peer_keepalive_timer; struct work_struct peer_keepalive_work; }; @@ -295,7 +295,7 @@ struct rxrpc_peer { struct hlist_head error_targets; /* targets for net error distribution */ struct work_struct error_distributor; struct rb_root service_conns; /* Service connections */ - struct hlist_node keepalive_link; /* Link in net->peer_keepalive[] */ + struct list_head keepalive_link; /* Link in net->peer_keepalive[] */ time64_t last_tx_at; /* Last time packet sent here */ seqlock_t service_conn_lock; spinlock_t lock; /* access lock */ diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c index 8229a52c2acd..3fde001fcc39 100644 --- a/net/rxrpc/conn_event.c +++ b/net/rxrpc/conn_event.c @@ -136,7 +136,7 @@ static void rxrpc_conn_retransmit_call(struct rxrpc_connection *conn, } ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, ioc, len); - conn->params.peer->last_tx_at = ktime_get_real(); + conn->params.peer->last_tx_at = ktime_get_seconds(); if (ret < 0) trace_rxrpc_tx_fail(conn->debug_id, serial, ret, rxrpc_tx_fail_call_final_resend); @@ -245,7 +245,7 @@ static int rxrpc_abort_connection(struct rxrpc_connection *conn, return -EAGAIN; } - conn->params.peer->last_tx_at = ktime_get_real(); + conn->params.peer->last_tx_at = ktime_get_seconds(); _leave(" = 0"); return 0; diff --git a/net/rxrpc/net_ns.c b/net/rxrpc/net_ns.c index 5d6a773db973..417d80867c4f 100644 --- a/net/rxrpc/net_ns.c +++ b/net/rxrpc/net_ns.c @@ -85,12 +85,12 @@ static __net_init int rxrpc_init_net(struct net *net) hash_init(rxnet->peer_hash); spin_lock_init(&rxnet->peer_hash_lock); for (i = 0; i < ARRAY_SIZE(rxnet->peer_keepalive); i++) - INIT_HLIST_HEAD(&rxnet->peer_keepalive[i]); - INIT_HLIST_HEAD(&rxnet->peer_keepalive_new); + INIT_LIST_HEAD(&rxnet->peer_keepalive[i]); + INIT_LIST_HEAD(&rxnet->peer_keepalive_new); timer_setup(&rxnet->peer_keepalive_timer, rxrpc_peer_keepalive_timeout, 0); INIT_WORK(&rxnet->peer_keepalive_work, rxrpc_peer_keepalive_worker); - rxnet->peer_keepalive_base = ktime_add(ktime_get_real(), NSEC_PER_SEC); + rxnet->peer_keepalive_base = ktime_get_seconds(); ret = -ENOMEM; rxnet->proc_net = proc_net_mkdir(net, "rxrpc", net->proc_net); diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c index f03de1c59ba3..4774c8f5634d 100644 --- a/net/rxrpc/output.c +++ b/net/rxrpc/output.c @@ -209,7 +209,7 @@ int rxrpc_send_ack_packet(struct rxrpc_call *call, bool ping, now = ktime_get_real(); if (ping) call->ping_time = now; - conn->params.peer->last_tx_at = ktime_get_real(); + conn->params.peer->last_tx_at = ktime_get_seconds(); if (ret < 0) trace_rxrpc_tx_fail(call->debug_id, serial, ret, rxrpc_tx_fail_call_ack); @@ -296,7 +296,7 @@ int rxrpc_send_abort_packet(struct rxrpc_call *call) ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 1, sizeof(pkt)); - conn->params.peer->last_tx_at = ktime_get_real(); + conn->params.peer->last_tx_at = ktime_get_seconds(); if (ret < 0) trace_rxrpc_tx_fail(call->debug_id, serial, ret, rxrpc_tx_fail_call_abort); @@ -391,7 +391,7 @@ int rxrpc_send_data_packet(struct rxrpc_call *call, struct sk_buff *skb, * message and update the peer record */ ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 2, len); - conn->params.peer->last_tx_at = ktime_get_real(); + conn->params.peer->last_tx_at = ktime_get_seconds(); up_read(&conn->params.local->defrag_sem); if (ret < 0) @@ -457,7 +457,7 @@ send_fragmentable: if (ret == 0) { ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 2, len); - conn->params.peer->last_tx_at = ktime_get_real(); + conn->params.peer->last_tx_at = ktime_get_seconds(); opt = IP_PMTUDISC_DO; kernel_setsockopt(conn->params.local->socket, SOL_IP, @@ -475,7 +475,7 @@ send_fragmentable: if (ret == 0) { ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 2, len); - conn->params.peer->last_tx_at = ktime_get_real(); + conn->params.peer->last_tx_at = ktime_get_seconds(); opt = IPV6_PMTUDISC_DO; kernel_setsockopt(conn->params.local->socket, @@ -599,6 +599,6 @@ void rxrpc_send_keepalive(struct rxrpc_peer *peer) trace_rxrpc_tx_fail(peer->debug_id, 0, ret, rxrpc_tx_fail_version_keepalive); - peer->last_tx_at = ktime_get_real(); + peer->last_tx_at = ktime_get_seconds(); _leave(""); } diff --git a/net/rxrpc/peer_event.c b/net/rxrpc/peer_event.c index 0ed8b651cec2..4f9da2f51c69 100644 --- a/net/rxrpc/peer_event.c +++ b/net/rxrpc/peer_event.c @@ -350,97 +350,117 @@ void rxrpc_peer_add_rtt(struct rxrpc_call *call, enum rxrpc_rtt_rx_trace why, } /* - * Perform keep-alive pings with VERSION packets to keep any NAT alive. + * Perform keep-alive pings. */ -void rxrpc_peer_keepalive_worker(struct work_struct *work) +static void rxrpc_peer_keepalive_dispatch(struct rxrpc_net *rxnet, + struct list_head *collector, + time64_t base, + u8 cursor) { - struct rxrpc_net *rxnet = - container_of(work, struct rxrpc_net, peer_keepalive_work); struct rxrpc_peer *peer; - unsigned long delay; - ktime_t base, now = ktime_get_real(); - s64 diff; - u8 cursor, slot; + const u8 mask = ARRAY_SIZE(rxnet->peer_keepalive) - 1; + time64_t keepalive_at; + int slot; - base = rxnet->peer_keepalive_base; - cursor = rxnet->peer_keepalive_cursor; + spin_lock_bh(&rxnet->peer_hash_lock); - _enter("%u,%lld", cursor, ktime_sub(now, base)); + while (!list_empty(collector)) { + peer = list_entry(collector->next, + struct rxrpc_peer, keepalive_link); -next_bucket: - diff = ktime_to_ns(ktime_sub(now, base)); - if (diff < 0) - goto resched; + list_del_init(&peer->keepalive_link); + if (!rxrpc_get_peer_maybe(peer)) + continue; - _debug("at %u", cursor); - spin_lock_bh(&rxnet->peer_hash_lock); -next_peer: - if (!rxnet->live) { spin_unlock_bh(&rxnet->peer_hash_lock); - goto out; - } - /* Everything in the bucket at the cursor is processed this second; the - * bucket at cursor + 1 goes now + 1s and so on... - */ - if (hlist_empty(&rxnet->peer_keepalive[cursor])) { - if (hlist_empty(&rxnet->peer_keepalive_new)) { - spin_unlock_bh(&rxnet->peer_hash_lock); - goto emptied_bucket; + keepalive_at = peer->last_tx_at + RXRPC_KEEPALIVE_TIME; + slot = keepalive_at - base; + _debug("%02x peer %u t=%d {%pISp}", + cursor, peer->debug_id, slot, &peer->srx.transport); + + if (keepalive_at <= base || + keepalive_at > base + RXRPC_KEEPALIVE_TIME) { + rxrpc_send_keepalive(peer); + slot = RXRPC_KEEPALIVE_TIME; } - hlist_move_list(&rxnet->peer_keepalive_new, - &rxnet->peer_keepalive[cursor]); + /* A transmission to this peer occurred since last we examined + * it so put it into the appropriate future bucket. + */ + slot += cursor; + slot &= mask; + spin_lock_bh(&rxnet->peer_hash_lock); + list_add_tail(&peer->keepalive_link, + &rxnet->peer_keepalive[slot & mask]); + rxrpc_put_peer(peer); } - peer = hlist_entry(rxnet->peer_keepalive[cursor].first, - struct rxrpc_peer, keepalive_link); - hlist_del_init(&peer->keepalive_link); - if (!rxrpc_get_peer_maybe(peer)) - goto next_peer; - spin_unlock_bh(&rxnet->peer_hash_lock); +} - _debug("peer %u {%pISp}", peer->debug_id, &peer->srx.transport); +/* + * Perform keep-alive pings with VERSION packets to keep any NAT alive. + */ +void rxrpc_peer_keepalive_worker(struct work_struct *work) +{ + struct rxrpc_net *rxnet = + container_of(work, struct rxrpc_net, peer_keepalive_work); + const u8 mask = ARRAY_SIZE(rxnet->peer_keepalive) - 1; + time64_t base, now, delay; + u8 cursor, stop; + LIST_HEAD(collector); -recalc: - diff = ktime_divns(ktime_sub(peer->last_tx_at, base), NSEC_PER_SEC); - if (diff < -30 || diff > 30) - goto send; /* LSW of 64-bit time probably wrapped on 32-bit */ - diff += RXRPC_KEEPALIVE_TIME - 1; - if (diff < 0) - goto send; + now = ktime_get_seconds(); + base = rxnet->peer_keepalive_base; + cursor = rxnet->peer_keepalive_cursor; + _enter("%lld,%u", base - now, cursor); - slot = (diff > RXRPC_KEEPALIVE_TIME - 1) ? RXRPC_KEEPALIVE_TIME - 1 : diff; - if (slot == 0) - goto send; + if (!rxnet->live) + return; - /* A transmission to this peer occurred since last we examined it so - * put it into the appropriate future bucket. + /* Remove to a temporary list all the peers that are currently lodged + * in expired buckets plus all new peers. + * + * Everything in the bucket at the cursor is processed this + * second; the bucket at cursor + 1 goes at now + 1s and so + * on... */ - slot = (slot + cursor) % ARRAY_SIZE(rxnet->peer_keepalive); spin_lock_bh(&rxnet->peer_hash_lock); - hlist_add_head(&peer->keepalive_link, &rxnet->peer_keepalive[slot]); - rxrpc_put_peer(peer); - goto next_peer; - -send: - rxrpc_send_keepalive(peer); - now = ktime_get_real(); - goto recalc; + list_splice_init(&rxnet->peer_keepalive_new, &collector); + + stop = cursor + ARRAY_SIZE(rxnet->peer_keepalive); + while (base <= now && (s8)(cursor - stop) < 0) { + list_splice_tail_init(&rxnet->peer_keepalive[cursor & mask], + &collector); + base++; + cursor++; + } -emptied_bucket: - cursor++; - if (cursor >= ARRAY_SIZE(rxnet->peer_keepalive)) - cursor = 0; - base = ktime_add_ns(base, NSEC_PER_SEC); - goto next_bucket; + base = now; + spin_unlock_bh(&rxnet->peer_hash_lock); -resched: rxnet->peer_keepalive_base = base; rxnet->peer_keepalive_cursor = cursor; - delay = nsecs_to_jiffies(-diff) + 1; - timer_reduce(&rxnet->peer_keepalive_timer, jiffies + delay); -out: + rxrpc_peer_keepalive_dispatch(rxnet, &collector, base, cursor); + ASSERT(list_empty(&collector)); + + /* Schedule the timer for the next occupied timeslot. */ + cursor = rxnet->peer_keepalive_cursor; + stop = cursor + RXRPC_KEEPALIVE_TIME - 1; + for (; (s8)(cursor - stop) < 0; cursor++) { + if (!list_empty(&rxnet->peer_keepalive[cursor & mask])) + break; + base++; + } + + now = ktime_get_seconds(); + delay = base - now; + if (delay < 1) + delay = 1; + delay *= HZ; + if (rxnet->live) + timer_reduce(&rxnet->peer_keepalive_timer, jiffies + delay); + _leave(""); } diff --git a/net/rxrpc/peer_object.c b/net/rxrpc/peer_object.c index 1b7e8107b3ae..24ec7cdcf332 100644 --- a/net/rxrpc/peer_object.c +++ b/net/rxrpc/peer_object.c @@ -322,7 +322,7 @@ struct rxrpc_peer *rxrpc_lookup_incoming_peer(struct rxrpc_local *local, if (!peer) { peer = prealloc; hash_add_rcu(rxnet->peer_hash, &peer->hash_link, hash_key); - hlist_add_head(&peer->keepalive_link, &rxnet->peer_keepalive_new); + list_add_tail(&peer->keepalive_link, &rxnet->peer_keepalive_new); } spin_unlock(&rxnet->peer_hash_lock); @@ -367,8 +367,8 @@ struct rxrpc_peer *rxrpc_lookup_peer(struct rxrpc_local *local, if (!peer) { hash_add_rcu(rxnet->peer_hash, &candidate->hash_link, hash_key); - hlist_add_head(&candidate->keepalive_link, - &rxnet->peer_keepalive_new); + list_add_tail(&candidate->keepalive_link, + &rxnet->peer_keepalive_new); } spin_unlock_bh(&rxnet->peer_hash_lock); @@ -441,7 +441,7 @@ static void __rxrpc_put_peer(struct rxrpc_peer *peer) spin_lock_bh(&rxnet->peer_hash_lock); hash_del_rcu(&peer->hash_link); - hlist_del_init(&peer->keepalive_link); + list_del_init(&peer->keepalive_link); spin_unlock_bh(&rxnet->peer_hash_lock); kfree_rcu(peer, rcu); diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c index 278ac0807a60..47cb019c521a 100644 --- a/net/rxrpc/rxkad.c +++ b/net/rxrpc/rxkad.c @@ -669,7 +669,7 @@ static int rxkad_issue_challenge(struct rxrpc_connection *conn) return -EAGAIN; } - conn->params.peer->last_tx_at = ktime_get_real(); + conn->params.peer->last_tx_at = ktime_get_seconds(); _leave(" = 0"); return 0; } @@ -725,7 +725,7 @@ static int rxkad_send_response(struct rxrpc_connection *conn, return -EAGAIN; } - conn->params.peer->last_tx_at = ktime_get_real(); + conn->params.peer->last_tx_at = ktime_get_seconds(); _leave(" = 0"); return 0; } -- cgit v1.2.3 From caa21e19e08d7a1445116a93f7ab4e187ebbbadb Mon Sep 17 00:00:00 2001 From: Ursula Braun Date: Wed, 8 Aug 2018 14:13:19 +0200 Subject: net/smc: no shutdown in state SMC_LISTEN Invoking shutdown for a socket in state SMC_LISTEN does not make sense. Nevertheless programs like syzbot fuzzing the kernel may try to do this. For SMC this means a socket refcounting problem. This patch makes sure a shutdown call for an SMC socket in state SMC_LISTEN simply returns with -ENOTCONN. Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/smc/af_smc.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'net') diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 05e4ffe5aabd..1288c7bf40d5 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -1397,8 +1397,7 @@ static int smc_shutdown(struct socket *sock, int how) lock_sock(sk); rc = -ENOTCONN; - if ((sk->sk_state != SMC_LISTEN) && - (sk->sk_state != SMC_ACTIVE) && + if ((sk->sk_state != SMC_ACTIVE) && (sk->sk_state != SMC_PEERCLOSEWAIT1) && (sk->sk_state != SMC_PEERCLOSEWAIT2) && (sk->sk_state != SMC_APPCLOSEWAIT1) && -- cgit v1.2.3 From bd58c7e0860f54710907903ed6daff699d1fc5b9 Mon Sep 17 00:00:00 2001 From: Ursula Braun Date: Wed, 8 Aug 2018 14:13:20 +0200 Subject: net/smc: allow sysctl rmem and wmem defaults for servers Without setsockopt SO_SNDBUF and SO_RCVBUF settings, the sysctl defaults net.ipv4.tcp_wmem and net.ipv4.tcp_rmem should be the base for the sizes of the SMC sndbuf and rcvbuf. Any TCP buffer size optimizations for servers should be ignored. Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/smc/af_smc.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'net') diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 1288c7bf40d5..0ee7721afbe5 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -1122,6 +1122,8 @@ static void smc_tcp_listen_work(struct work_struct *work) sock_hold(lsk); /* sock_put in smc_listen_work */ INIT_WORK(&new_smc->smc_listen_work, smc_listen_work); smc_copy_sock_settings_to_smc(new_smc); + new_smc->sk.sk_sndbuf = lsmc->sk.sk_sndbuf; + new_smc->sk.sk_rcvbuf = lsmc->sk.sk_rcvbuf; sock_hold(&new_smc->sk); /* sock_put in passive closing */ if (!schedule_work(&new_smc->smc_listen_work)) sock_put(&new_smc->sk); -- cgit v1.2.3 From 7311d665ca68907b9c43d6d1021f816f9a7bbd57 Mon Sep 17 00:00:00 2001 From: Ursula Braun Date: Wed, 8 Aug 2018 14:13:21 +0200 Subject: net/smc: move sock lock in smc_ioctl() When an SMC socket is connecting it is decided whether fallback to TCP is needed. To avoid races between connect and ioctl move the sock lock before the use_fallback check. Reported-by: syzbot+5b2cece1a8ecb2ca77d8@syzkaller.appspotmail.com Reported-by: syzbot+19557374321ca3710990@syzkaller.appspotmail.com Fixes: 1992d99882af ("net/smc: take sock lock in smc_ioctl()") Signed-off-by: Ursula Braun Signed-off-by: David S. Miller --- net/smc/af_smc.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'net') diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index 0ee7721afbe5..e7de5f282722 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -1522,12 +1522,16 @@ static int smc_ioctl(struct socket *sock, unsigned int cmd, smc = smc_sk(sock->sk); conn = &smc->conn; + lock_sock(&smc->sk); if (smc->use_fallback) { - if (!smc->clcsock) + if (!smc->clcsock) { + release_sock(&smc->sk); return -EBADF; - return smc->clcsock->ops->ioctl(smc->clcsock, cmd, arg); + } + answ = smc->clcsock->ops->ioctl(smc->clcsock, cmd, arg); + release_sock(&smc->sk); + return answ; } - lock_sock(&smc->sk); switch (cmd) { case SIOCINQ: /* same as FIONREAD */ if (smc->sk.sk_state == SMC_LISTEN) { -- cgit v1.2.3 From 1be52e97ed3e524f82e25d6e53f48df3c6e85282 Mon Sep 17 00:00:00 2001 From: Andrew Lunn Date: Wed, 8 Aug 2018 20:56:40 +0200 Subject: dsa: slave: eee: Allow ports to use phylink For a port to be able to use EEE, both the MAC and the PHY must support EEE. A phy can be provided by both a phydev or phylink. Verify at least one of these exist, not just phydev. Fixes: aab9c4067d23 ("net: dsa: Plug in PHYLINK support") Signed-off-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- net/dsa/slave.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'net') diff --git a/net/dsa/slave.c b/net/dsa/slave.c index 732369c80644..9864bcd3d317 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -639,7 +639,7 @@ static int dsa_slave_set_eee(struct net_device *dev, struct ethtool_eee *e) int ret; /* Port's PHY and MAC both need to be EEE capable */ - if (!dev->phydev) + if (!dev->phydev && !dp->pl) return -ENODEV; if (!ds->ops->set_mac_eee) @@ -659,7 +659,7 @@ static int dsa_slave_get_eee(struct net_device *dev, struct ethtool_eee *e) int ret; /* Port's PHY and MAC both need to be EEE capable */ - if (!dev->phydev) + if (!dev->phydev && !dp->pl) return -ENODEV; if (!ds->ops->get_mac_eee) -- cgit v1.2.3