From 3f32d0be6c16b902b687453c962d17eea5b8ea19 Mon Sep 17 00:00:00 2001 From: Parthasarathy Bhuvaragan Date: Tue, 25 Sep 2018 22:09:10 +0200 Subject: tipc: lock wakeup & inputq at tipc_link_reset() In tipc_link_reset() we copy the wakeup queue to input queue using skb_queue_splice_init(link->wakeupq, link->inputq). This is performed without holding any locks. The lists might be simultaneously be accessed by other cpu threads in tipc_sk_rcv(), something leading to to random missing packets. Signed-off-by: Parthasarathy Bhuvaragan Signed-off-by: Jon Maloy Signed-off-by: David S. Miller --- net/tipc/link.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'net/tipc/link.c') diff --git a/net/tipc/link.c b/net/tipc/link.c index b1f0bee54eac..26cc033ee167 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -841,9 +841,14 @@ void tipc_link_reset(struct tipc_link *l) l->in_session = false; l->session++; l->mtu = l->advertised_mtu; + spin_lock_bh(&l->wakeupq.lock); + spin_lock_bh(&l->inputq->lock); + skb_queue_splice_init(&l->wakeupq, l->inputq); + spin_unlock_bh(&l->inputq->lock); + spin_unlock_bh(&l->wakeupq.lock); + __skb_queue_purge(&l->transmq); __skb_queue_purge(&l->deferdq); - skb_queue_splice_init(&l->wakeupq, l->inputq); __skb_queue_purge(&l->backlogq); l->backlog[TIPC_LOW_IMPORTANCE].len = 0; l->backlog[TIPC_MEDIUM_IMPORTANCE].len = 0; -- cgit v1.2.3 From c140eb166d681f66bd7e99fb121357db1a503e7f Mon Sep 17 00:00:00 2001 From: LUU Duc Canh Date: Wed, 26 Sep 2018 21:00:54 +0200 Subject: tipc: fix failover problem We see the following scenario: 1) Link endpoint B on node 1 discovers that its peer endpoint is gone. Since there is a second working link, failover procedure is started. 2) Link endpoint A on node 1 sends a FAILOVER message to peer endpoint A on node 2. The node item 1->2 goes to state FAILINGOVER. 3) Linke endpoint A/2 receives the failover, and is supposed to take down its parallell link endpoint B/2, while producing a FAILOVER message to send back to A/1. 4) However, B/2 has already been deleted, so no FAILOVER message can created. 5) Node 1->2 remains in state FAILINGOVER forever, refusing to receive any messages that can bring B/1 up again. We are left with a non- redundant link between node 1 and 2. We fix this with letting endpoint A/2 build a dummy FAILOVER message to send to back to A/1, so that the situation can be resolved. Signed-off-by: LUU Duc Canh Signed-off-by: Jon Maloy Signed-off-by: David S. Miller --- net/tipc/link.c | 35 +++++++++++++++++++++++++++++++++++ 1 file changed, 35 insertions(+) (limited to 'net/tipc/link.c') diff --git a/net/tipc/link.c b/net/tipc/link.c index 26cc033ee167..4ed650ce6e61 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -410,6 +410,11 @@ char *tipc_link_name(struct tipc_link *l) return l->name; } +u32 tipc_link_state(struct tipc_link *l) +{ + return l->state; +} + /** * tipc_link_create - create a new link * @n: pointer to associated node @@ -1385,6 +1390,36 @@ static void tipc_link_build_proto_msg(struct tipc_link *l, int mtyp, bool probe, __skb_queue_tail(xmitq, skb); } +void tipc_link_create_dummy_tnl_msg(struct tipc_link *l, + struct sk_buff_head *xmitq) +{ + u32 onode = tipc_own_addr(l->net); + struct tipc_msg *hdr, *ihdr; + struct sk_buff_head tnlq; + struct sk_buff *skb; + u32 dnode = l->addr; + + skb_queue_head_init(&tnlq); + skb = tipc_msg_create(TUNNEL_PROTOCOL, FAILOVER_MSG, + INT_H_SIZE, BASIC_H_SIZE, + dnode, onode, 0, 0, 0); + if (!skb) { + pr_warn("%sunable to create tunnel packet\n", link_co_err); + return; + } + + hdr = buf_msg(skb); + msg_set_msgcnt(hdr, 1); + msg_set_bearer_id(hdr, l->peer_bearer_id); + + ihdr = (struct tipc_msg *)msg_data(hdr); + tipc_msg_init(onode, ihdr, TIPC_LOW_IMPORTANCE, TIPC_DIRECT_MSG, + BASIC_H_SIZE, dnode); + msg_set_errcode(ihdr, TIPC_ERR_NO_PORT); + __skb_queue_tail(&tnlq, skb); + tipc_link_xmit(l, &tnlq, xmitq); +} + /* tipc_link_tnl_prepare(): prepare and return a list of tunnel packets * with contents of the link's transmit and backlog queues. */ -- cgit v1.2.3 From d949cfedbcbab4e91590576cbace2671924ad69c Mon Sep 17 00:00:00 2001 From: LUU Duc Canh Date: Wed, 26 Sep 2018 22:28:52 +0200 Subject: tipc: ignore STATE_MSG on wrong link session The initial session number when a link is created is based on a random value, taken from struct tipc_net->random. It is then incremented for each link reset to avoid mixing protocol messages from different link sessions. However, when a bearer is reset all its links are deleted, and will later be re-created using the same random value as the first time. This means that if the link never went down between creation and deletion we will still sometimes have two subsequent sessions with the same session number. In virtual environments with potentially long transmission times this has turned out to be a real problem. We now fix this by randomizing the session number each time a link is created. With a session number size of 16 bits this gives a risk of session collision of 1/64k. To reduce this further, we also introduce a sanity check on the very first STATE message arriving at a link. If this has an acknowledge value differing from 0, which is logically impossible, we ignore the message. The final risk for session collision is hence reduced to 1/4G, which should be sufficient. Signed-off-by: LUU Duc Canh Signed-off-by: Jon Maloy Signed-off-by: David S. Miller --- net/tipc/link.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'net/tipc/link.c') diff --git a/net/tipc/link.c b/net/tipc/link.c index 4ed650ce6e61..fb886b525d95 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -1516,6 +1516,9 @@ bool tipc_link_validate_msg(struct tipc_link *l, struct tipc_msg *hdr) return false; if (session != curr_session) return false; + /* Extra sanity check */ + if (!link_is_up(l) && msg_ack(hdr)) + return false; if (!(l->peer_caps & TIPC_LINK_PROTO_SEQNO)) return true; /* Accept only STATE with new sequence number */ -- cgit v1.2.3