From ed2e923945892a8372ab70d2f61d364b0b6d9054 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 19 Sep 2015 09:08:34 -0700 Subject: tcp/dccp: fix timewait races in timer handling When creating a timewait socket, we need to arm the timer before allowing other cpus to find it. The signal allowing cpus to find the socket is setting tw_refcnt to non zero value. As we set tw_refcnt in __inet_twsk_hashdance(), we therefore need to call inet_twsk_schedule() first. This also means we need to remove tw_refcnt changes from inet_twsk_schedule() and let the caller handle it. Note that because we use mod_timer_pinned(), we have the guarantee the timer wont expire before we set tw_refcnt as we run in BH context. To make things more readable I introduced inet_twsk_reschedule() helper. When rearming the timer, we can use mod_timer_pending() to make sure we do not rearm a canceled timer. Note: This bug can possibly trigger if packets of a flow can hit multiple cpus. This does not normally happen, unless flow steering is broken somehow. This explains this bug was spotted ~5 months after its introduction. A similar fix is needed for SYN_RECV sockets in reqsk_queue_hash_req(), but will be provided in a separate patch for proper tracking. Fixes: 789f558cfb36 ("tcp/dccp: get rid of central timewait timer") Signed-off-by: Eric Dumazet Reported-by: Ying Cai Signed-off-by: David S. Miller --- net/ipv4/inet_timewait_sock.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) (limited to 'net/ipv4/inet_timewait_sock.c') diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c index ae22cc24fbe8..c67f9bd7699c 100644 --- a/net/ipv4/inet_timewait_sock.c +++ b/net/ipv4/inet_timewait_sock.c @@ -123,13 +123,15 @@ void __inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk, /* * Step 2: Hash TW into tcp ehash chain. * Notes : - * - tw_refcnt is set to 3 because : + * - tw_refcnt is set to 4 because : * - We have one reference from bhash chain. * - We have one reference from ehash chain. + * - We have one reference from timer. + * - One reference for ourself (our caller will release it). * We can use atomic_set() because prior spin_lock()/spin_unlock() * committed into memory all tw fields. */ - atomic_set(&tw->tw_refcnt, 1 + 1 + 1); + atomic_set(&tw->tw_refcnt, 4); inet_twsk_add_node_rcu(tw, &ehead->chain); /* Step 3: Remove SK from hash chain */ @@ -217,7 +219,7 @@ void inet_twsk_deschedule_put(struct inet_timewait_sock *tw) } EXPORT_SYMBOL(inet_twsk_deschedule_put); -void inet_twsk_schedule(struct inet_timewait_sock *tw, const int timeo) +void __inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo, bool rearm) { /* timeout := RTO * 3.5 * @@ -245,12 +247,14 @@ void inet_twsk_schedule(struct inet_timewait_sock *tw, const int timeo) */ tw->tw_kill = timeo <= 4*HZ; - if (!mod_timer_pinned(&tw->tw_timer, jiffies + timeo)) { - atomic_inc(&tw->tw_refcnt); + if (!rearm) { + BUG_ON(mod_timer_pinned(&tw->tw_timer, jiffies + timeo)); atomic_inc(&tw->tw_dr->tw_count); + } else { + mod_timer_pending(&tw->tw_timer, jiffies + timeo); } } -EXPORT_SYMBOL_GPL(inet_twsk_schedule); +EXPORT_SYMBOL_GPL(__inet_twsk_schedule); void inet_twsk_purge(struct inet_hashinfo *hashinfo, struct inet_timewait_death_row *twdr, int family) -- cgit v1.2.3-59-g8ed1b