netfilter: nf_conntrack: don't release a conntrack with non-zero refcnt
Subsystems (2)
Pablo Neira Ayuso
maintainerPATCH 1/1 netfilter: nf_conntrack: don't release a conntrack with non-zero refcnt
With this patch, the conntrack refcount is initially set to zero and it is bumped once it is added to any of the list, so we fulfill Eric's golden rule which is that all released objects always have a refcount that equals zero.
Andrey Vagin reports that nf_conntrack_free can't be called for a conntrack with non-zero ref-counter, because it can race with nf_conntrack_find_get().
A conntrack slab is created with SLAB_DESTROY_BY_RCU. Non-zero ref-counter says that this conntrack is used. So when we release a conntrack with non-zero counter, we break this assumption.
CPU1 CPU2____nf_conntrack_find()
nf_ct_put()
destroy_conntrack()
...
init_conntrack
__nf_conntrack_alloc (set use = 1)
atomic_inc_not_zero(&ct->use) (use = 2)
if (!l4proto->new(ct, skb, dataoff, timeouts))
nf_conntrack_free(ct); (use = 2 !!!)
...
__nf_conntrack_alloc (set use = 1)
if (!nf_ct_key_equal(h, tuple, zone))
nf_ct_put(ct); (use = 0)
destroy_conntrack()
/* continue to work with CT */After applying the path "[PATCH] netfilter: nf_conntrack: fix RCU race in nf_conntrack_find_get" another bug was triggered in destroy_conntrack():
<4>[67096.759334] ------------[ cut here ]------------ <2>[67096.759353] kernel BUG at net/netfilter/nf_conntrack_core.c:211! ... <4>[67096.759837] Pid: 498649, comm: atdd veid: 666 Tainted: G C --------------- 2.6.32-042stab084.18 #1 042stab084_18 /DQ45CB <4>[67096.759932] RIP: 0010:[<ffffffffa03d99ac>] [<ffffffffa03d99ac>] destroy_conntrack+0x15c/0x190 [nf_conntrack] <4>[67096.760255] Call Trace: <4>[67096.760255] [<ffffffff814844a7>] nf_conntrack_destroy+0x17/0x30 <4>[67096.760255] [<ffffffffa03d9bb5>] nf_conntrack_find_get+0x85/0x130 [nf_conntrack] <4>[67096.760255] [<ffffffffa03d9fb2>] nf_conntrack_in+0x352/0xb60 [nf_conntrack] <4>[67096.760255] [<ffffffffa048c771>] ipv4_conntrack_local+0x51/0x60 [nf_conntrack_ipv4] <4>[67096.760255] [<ffffffff81484419>] nf_iterate+0x69/0xb0 <4>[67096.760255] [<ffffffff814b5b00>] ? dst_output+0x0/0x20 <4>[67096.760255] [<ffffffff814845d4>] nf_hook_slow+0x74/0x110 <4>[67096.760255] [<ffffffff814b5b00>] ? dst_output+0x0/0x20 <4>[67096.760255] [<ffffffff814b66d5>] raw_sendmsg+0x775/0x910 <4>[67096.760255] [<ffffffff8104c5a8>] ? flush_tlb_others_ipi+0x128/0x130 <4>[67096.760255] [<ffffffff8100bc4e>] ? apic_timer_interrupt+0xe/0x20 <4>[67096.760255] [<ffffffff8100bc4e>] ? apic_timer_interrupt+0xe/0x20 <4>[67096.760255] [<ffffffff814c136a>] inet_sendmsg+0x4a/0xb0 <4>[67096.760255] [<ffffffff81444e93>] ? sock_sendmsg+0x13/0x140 <4>[67096.760255] [<ffffffff81444f97>] sock_sendmsg+0x117/0x140 <4>[67096.760255] [<ffffffff8102e299>] ? native_smp_send_reschedule+0x49/0x60 <4>[67096.760255] [<ffffffff81519beb>] ? _spin_unlock_bh+0x1b/0x20 <4>[67096.760255] [<ffffffff8109d930>] ? autoremove_wake_function+0x0/0x40 <4>[67096.760255] [<ffffffff814960f0>] ? do_ip_setsockopt+0x90/0xd80 <4>[67096.760255] [<ffffffff8100bc4e>] ? apic_timer_interrupt+0xe/0x20 <4>[67096.760255] [<ffffffff8100bc4e>] ? apic_timer_interrupt+0xe/0x20 <4>[67096.760255] [<ffffffff814457c9>] sys_sendto+0x139/0x190 <4>[67096.760255] [<ffffffff810efa77>] ? audit_syscall_entry+0x1d7/0x200 <4>[67096.760255] [<ffffffff810ef7c5>] ? __audit_syscall_exit+0x265/0x290 <4>[67096.760255] [<ffffffff81474daf>] compat_sys_socketcall+0x13f/0x210 <4>[67096.760255] [<ffffffff8104dea3>] ia32_sysret+0x0/0x5
---
I have reused the original title for the RFC patch that Andrey posted and most of the original patch description.
Thanks to Eric Dumazet for indicating changes regarding memory barriers.
include/net/netfilter/nf_conntrack.h | 2 ++ net/netfilter/nf_conntrack_core.c | 34 +++++++++++++++++++++++++++++----- net/netfilter/nf_synproxy_core.c | 5 ++--- net/netfilter/xt_CT.c | 7 +------ 4 files changed, 34 insertions(+), 14 deletions(-)
include/net/netfilter/nf_conntrack.h +2 -0
@@ -284,6 +284,8 @@ extern unsigned int nf_conntrack_max;extern unsigned int nf_conntrack_hash_rnd;void init_nf_conntrack_hash_rnd (void ); +void nf_conntrack_tmpl_insert (struct net *net,struct nf_conn *tmpl); +#define NF_CT_STAT_INC (net, count) __this_cpu_inc((net)->ct.stat->count)#define NF_CT_STAT_INC_ATOMIC (net, count) this_cpu_inc((net)->ct.stat->count)
net/netfilter/nf_conntrack_core.c +29 -5
@@ -448,7 +448,9 @@ nf_conntrack_hash_check_insert(struct nf_conn *ct) goto out;add_timer (&ct->timeout); -nf_conntrack_get (&ct->ct_general); +smp_wmb (); +/* The caller holds a reference to this object */ +atomic_set (&ct->ct_general.use,2 );__nf_conntrack_hash_insert (ct, hash, repl_hash);NF_CT_STAT_INC (net, insert);spin_unlock_bh (&nf_conntrack_lock);
@@ -462,6 +464,21 @@ out: }EXPORT_SYMBOL_GPL (nf_conntrack_hash_check_insert); +/* deletion from this larval template list happens via nf_ct_put() */ +void nf_conntrack_tmpl_insert (struct net *net,struct nf_conn *tmpl) +{ +__set_bit (IPS_TEMPLATE_BIT , &tmpl->status); +__set_bit (IPS_CONFIRMED_BIT , &tmpl->status); +nf_conntrack_get (&tmpl->ct_general); + +spin_lock_bh (&nf_conntrack_lock); +/* Overload tuple linked list to put us in template list. */ +hlist_nulls_add_head_rcu (&tmpl->tuplehash[IP_CT_DIR_ORIGINAL ].hnnode, + &net->ct.tmpl); +spin_unlock_bh (&nf_conntrack_lock); +} +EXPORT_SYMBOL_GPL (nf_conntrack_tmpl_insert); +/* Confirm a connection given skb; places it in hash table */ int __nf_conntrack_confirm (struct sk_buff *skb)
@@ -733,11 +750,10 @@ __nf_conntrack_alloc(struct net *net, u16 zone, nf_ct_zone->id = zone; }#endif -/* -* changes to lookup keys must be done before setting refcnt to 1 +/* Because we use RCU lookups, we set ct_general.use to zero before +* this is inserted in any list. */ -smp_wmb (); -atomic_set (&ct->ct_general.use,1 ); +atomic_set (&ct->ct_general.use,0 );return ct;#ifdef CONFIG_NF_CONNTRACK_ZONES
@@ -761,6 +777,11 @@ void nf_conntrack_free(struct nf_conn *ct) {struct net *net =nf_ct_net (ct); +/* A freed object has refcnt == 0, that's +* the golden rule for SLAB_DESTROY_BY_RCU +*/ +NF_CT_ASSERT (atomic_read (&ct->ct_general.use) ==0 ); +nf_ct_ext_destroy (ct);nf_ct_ext_free (ct);kmem_cache_free (net->ct.nf_conntrack_cachep, ct);
@@ -856,6 +877,9 @@ init_conntrack(struct net *net, struct nf_conn *tmpl,NF_CT_STAT_INC (net, new); } +/* Now it is inserted into the unconfirmed list, bump refcount */ +nf_conntrack_get (&ct->ct_general); +/* Overload tuple linked list to put us in unconfirmed list. */ hlist_nulls_add_head_rcu (&ct->tuplehash[IP_CT_DIR_ORIGINAL ].hnnode, &net->ct.unconfirmed);
net/netfilter/nf_synproxy_core.c +2 -3
@@ -363,9 +363,8 @@ static int __net_init synproxy_net_init(struct net *net) goto err2;if (!nfct_synproxy_ext_add (ct)) goto err2; -__set_bit (IPS_TEMPLATE_BIT , &ct->status); -__set_bit (IPS_CONFIRMED_BIT , &ct->status); +nf_conntrack_tmpl_insert (net, ct); snet->tmpl = ct; snet->stats = alloc_percpu(struct synproxy_stats);
@@ -390,7 +389,7 @@ static void __net_exit synproxy_net_exit(struct net *net) {struct synproxy_net *snet =synproxy_pernet (net); -nf_conntrack_free (snet->tmpl); +nf_ct_put (snet->tmpl);synproxy_proc_exit (net);free_percpu (snet->stats); }
net/netfilter/xt_CT.c +1 -6
@@ -228,12 +228,7 @@ static int xt_ct_tg_check(const struct xt_tgchk_param *par, goto err3; } -__set_bit (IPS_TEMPLATE_BIT , &ct->status); -__set_bit (IPS_CONFIRMED_BIT , &ct->status); - -/* Overload tuple linked list to put us in template list. */ -hlist_nulls_add_head_rcu (&ct->tuplehash[IP_CT_DIR_ORIGINAL ].hnnode, - &par->net->ct.tmpl); +nf_conntrack_tmpl_insert (par->net, ct); out: info->ct = ct;return 0 ;
Signature · 1 line
1.7.10.4
Florian Westphal
maintainernet/netfilter/nf_conntrack_core.c nf_conntrack_hash_check_insert(struct nf_conn *ct) · line 453
Quoted 12 lines · Pablo Neira Ayuso
net/netfilter/nf_conntrack_core.c +3
-1@@ -448,7 +448,9 @@ nf_conntrack_hash_check_insert(struct nf_conn *ct) goto out;add_timer (&ct->timeout); -nf_conntrack_get (&ct->ct_general); +smp_wmb (); +/* The caller holds a reference to this object */ +atomic_set (&ct->ct_general.use,2 );
What happens when the skb is dropped before confirmation? How is skb_clone etc. solved (it increments refcnt of underlying conntrack object?)
Pablo Neira Ayuso
maintainerQuoted 8 lines · Florian Westphal
Quoted 7 lines · Pablo Neira Ayuso
net/netfilter/nf_conntrack_core.c +0
-0@@ -448,7 +448,9 @@ nf_conntrack_hash_check_insert(struct nf_conn *ct) goto out;
Quoted 7 lines
net/netfilter/nf_conntrack_core.c nf_conntrack_hash_check_insert(struct nf_conn *ct) · line 453
Quoted 5 lines
net/netfilter/nf_conntrack_core.c +3
-1add_timer (&ct->timeout); -nf_conntrack_get (&ct->ct_general); +smp_wmb (); +/* The caller holds a reference to this object */ +atomic_set (&ct->ct_general.use,2 );What happens when the skb is dropped before confirmation?
If you refer to the spot above, conntracks added via ctnetlink are always confirmed and the refcount will become 1 after insertion into the hashes. In the packet path, the refcount is 1 for unconfirmed conntracks, so if you release the skbuff, nf_conntrack_put() is called and the conntrack is also released (fulfilling that refcount equals 0).
Quoted 2 lines
How is
skb_cloneetc. solved (it increments refcnt of underlying conntrack object?)
I think this works as before, this patch is just delaying the initial refcount bumping to 1 to make it once the conntrack is inserted in any of the lists.
Please, let me know if you still see any problematic case. Thanks.
Florian Westphal
maintainerQuoted 17 lines · Pablo Neira Ayuso
Quoted 8 lines · Florian Westphal
Quoted 7 lines · Pablo Neira Ayuso
net/netfilter/nf_conntrack_core.c +0
-0@@ -448,7 +448,9 @@ nf_conntrack_hash_check_insert(struct nf_conn *ct) goto out;Quoted 7 lines
net/netfilter/nf_conntrack_core.c nf_conntrack_hash_check_insert(struct nf_conn *ct) · line 453
Quoted 5 lines
net/netfilter/nf_conntrack_core.c +3
-1add_timer (&ct->timeout); -nf_conntrack_get (&ct->ct_general); +smp_wmb (); +/* The caller holds a reference to this object */ +atomic_set (&ct->ct_general.use,2 );What happens when the skb is dropped before confirmation?
Quoted 4 lines
If you refer to the spot above, conntracks added via ctnetlink are always confirmed and the refcount will become 1 after insertion into the hashes. In the packet path, the refcount is 1 for unconfirmed conntracks,
Right, sorry. I missed this. You only want to delay the init-to-1 until the conntrack is inserted into the unconfirmed list.
However, I think there is still a problem with the atomic_set to 2 above.
What about e.g.
Prerouting -> set refcnt to 1
(clone)
(clone)
(refcnt is 3)
forward/postrouting -> confirmation (refcnt reset to two)
I think this could happen at least with netfilter on-top-of bridge when dealing with e.g. udp mcast. (i am looking at br_forward and deliver_clone in particular).
I think you could just leave hash_check_insert() alone?
Pablo Neira Ayuso
maintainerQuoted 36 lines · Florian Westphal
Quoted 17 lines · Pablo Neira Ayuso
Quoted 8 lines · Florian Westphal
Quoted 7 lines · Pablo Neira Ayuso
net/netfilter/nf_conntrack_core.c +0
-0@@ -448,7 +448,9 @@ nf_conntrack_hash_check_insert(struct nf_conn *ct) goto out;Quoted 7 lines
net/netfilter/nf_conntrack_core.c nf_conntrack_hash_check_insert(struct nf_conn *ct) · line 453
Quoted 5 lines
net/netfilter/nf_conntrack_core.c +3
-1add_timer (&ct->timeout); -nf_conntrack_get (&ct->ct_general); +smp_wmb (); +/* The caller holds a reference to this object */ +atomic_set (&ct->ct_general.use,2 );What happens when the skb is dropped before confirmation?
Quoted 4 lines
If you refer to the spot above, conntracks added via ctnetlink are always confirmed and the refcount will become 1 after insertion into the hashes. In the packet path, the refcount is 1 for unconfirmed conntracks,
Right, sorry. I missed this. You only want to delay the init-to-1 until the conntrack is inserted into the unconfirmed list.
However, I think there is still a problem with the
atomic_setto 2 above.What about e.g.
Prerouting -> set refcnt to 1
(clone)
(clone)
(refcnt is 3)
forward/postrouting -> confirmation (refcnt reset to two)
Quoted 5 lines
I think this could happen at least with netfilter on-top-of bridge when dealing with e.g. udp mcast. (i am looking at
br_forwardanddeliver_clonein particular).I think you could just leave
hash_check_insert()alone?
hash_check_insert() is only used by ctnetlink in the creation path for new conntracks.
Florian Westphal
maintainerQuoted 4 lines · Pablo Neira Ayuso
Quoted 1 line
I think you could just leave
hash_check_insert()alone?
hash_check_insert()is only used by ctnetlink in the creation path for new conntracks.
I see. In that case I see nothing wrong with your patch.
Thanks for explaining!
Quoted 12 lines · Pablo Neira Ayuso
With this patch, the conntrack refcount is initially set to zero and it is bumped once it is added to any of the list, so we fulfill Eric's golden rule which is that all released objects always have a refcount that equals zero.
Andrey Vagin reports that
nf_conntrack_freecan't be called for a conntrack with non-zero ref-counter, because it can race withnf_conntrack_find_get().A conntrack slab is created with
SLAB_DESTROY_BY_RCU. Non-zero ref-counter says that this conntrack is used. So when we release a conntrack with non-zero counter, we break this assumption.
...
Quoted 6 lines
Cc Eric Dumazet <[email protected]>Cc Florian Westphal <[email protected]>Cc Andrew Vagin <[email protected]>Reported-by Andrew Vagin <[email protected]>Signed-off-by Pablo Neira Ayuso <[email protected]>---
SGTM !
Quoted 21 lines · Eric Dumazet
Quoted 12 lines · Pablo Neira Ayuso
With this patch, the conntrack refcount is initially set to zero and it is bumped once it is added to any of the list, so we fulfill Eric's golden rule which is that all released objects always have a refcount that equals zero.
Andrey Vagin reports that
nf_conntrack_freecan't be called for a conntrack with non-zero ref-counter, because it can race withnf_conntrack_find_get().A conntrack slab is created with
SLAB_DESTROY_BY_RCU. Non-zero ref-counter says that this conntrack is used. So when we release a conntrack with non-zero counter, we break this assumption....
Quoted 6 lines
Cc Eric Dumazet <[email protected]>Cc Florian Westphal <[email protected]>Cc Andrew Vagin <[email protected]>Reported-by Andrew Vagin <[email protected]>Signed-off-by Pablo Neira Ayuso <[email protected]>---
Quoted 3 lines
SGTM !
Reviewed-by Eric Dumazet <[email protected]>
Applied, thanks everyone!
Quoted 67 lines · Pablo Neira Ayuso
With this patch, the conntrack refcount is initially set to zero and it is bumped once it is added to any of the list, so we fulfill Eric's golden rule which is that all released objects always have a refcount that equals zero.
Andrey Vagin reports that
nf_conntrack_freecan't be called for a conntrack with non-zero ref-counter, because it can race withnf_conntrack_find_get().A conntrack slab is created with
SLAB_DESTROY_BY_RCU. Non-zero ref-counter says that this conntrack is used. So when we release a conntrack with non-zero counter, we break this assumption.CPU1 CPU2
____nf_conntrack_find()nf_ct_put() destroy_conntrack() ... init_conntrack __nf_conntrack_alloc (set use = 1) atomic_inc_not_zero(&ct->use) (use = 2) if (!l4proto->new(ct, skb, dataoff, timeouts)) nf_conntrack_free(ct); (use = 2 !!!) ... __nf_conntrack_alloc (set use = 1) if (!nf_ct_key_equal(h, tuple, zone)) nf_ct_put(ct); (use = 0) destroy_conntrack() /* continue to work with CT */After applying the path "[PATCH] netfilter:
nf_conntrack: fix RCU race innf_conntrack_find_get" another bug was triggered indestroy_conntrack():<4>[67096.759334] ------------[ cut here ]------------ <2>[67096.759353] kernel BUG at net/netfilter/nf_conntrack_core.c:211! ... <4>[67096.759837] Pid: 498649, comm: atdd veid: 666 Tainted: G C --------------- 2.6.32-042stab084.18 #1 042stab084_18 /DQ45CB <4>[67096.759932] RIP: 0010:[<ffffffffa03d99ac>] [<ffffffffa03d99ac>] destroy_conntrack+0x15c/0x190 [nf_conntrack] <4>[67096.760255] Call Trace: <4>[67096.760255] [<ffffffff814844a7>] nf_conntrack_destroy+0x17/0x30 <4>[67096.760255] [<ffffffffa03d9bb5>] nf_conntrack_find_get+0x85/0x130 [nf_conntrack] <4>[67096.760255] [<ffffffffa03d9fb2>] nf_conntrack_in+0x352/0xb60 [nf_conntrack] <4>[67096.760255] [<ffffffffa048c771>] ipv4_conntrack_local+0x51/0x60 [nf_conntrack_ipv4] <4>[67096.760255] [<ffffffff81484419>] nf_iterate+0x69/0xb0 <4>[67096.760255] [<ffffffff814b5b00>] ? dst_output+0x0/0x20 <4>[67096.760255] [<ffffffff814845d4>] nf_hook_slow+0x74/0x110 <4>[67096.760255] [<ffffffff814b5b00>] ? dst_output+0x0/0x20 <4>[67096.760255] [<ffffffff814b66d5>] raw_sendmsg+0x775/0x910 <4>[67096.760255] [<ffffffff8104c5a8>] ? flush_tlb_others_ipi+0x128/0x130 <4>[67096.760255] [<ffffffff8100bc4e>] ? apic_timer_interrupt+0xe/0x20 <4>[67096.760255] [<ffffffff8100bc4e>] ? apic_timer_interrupt+0xe/0x20 <4>[67096.760255] [<ffffffff814c136a>] inet_sendmsg+0x4a/0xb0 <4>[67096.760255] [<ffffffff81444e93>] ? sock_sendmsg+0x13/0x140 <4>[67096.760255] [<ffffffff81444f97>] sock_sendmsg+0x117/0x140 <4>[67096.760255] [<ffffffff8102e299>] ? native_smp_send_reschedule+0x49/0x60 <4>[67096.760255] [<ffffffff81519beb>] ? _spin_unlock_bh+0x1b/0x20 <4>[67096.760255] [<ffffffff8109d930>] ? autoremove_wake_function+0x0/0x40 <4>[67096.760255] [<ffffffff814960f0>] ? do_ip_setsockopt+0x90/0xd80 <4>[67096.760255] [<ffffffff8100bc4e>] ? apic_timer_interrupt+0xe/0x20 <4>[67096.760255] [<ffffffff8100bc4e>] ? apic_timer_interrupt+0xe/0x20 <4>[67096.760255] [<ffffffff814457c9>] sys_sendto+0x139/0x190 <4>[67096.760255] [<ffffffff810efa77>] ? audit_syscall_entry+0x1d7/0x200 <4>[67096.760255] [<ffffffff810ef7c5>] ? __audit_syscall_exit+0x265/0x290 <4>[67096.760255] [<ffffffff81474daf>] compat_sys_socketcall+0x13f/0x210 <4>[67096.760255] [<ffffffff8104dea3>] ia32_sysret+0x0/0x5
Quoted 3 lines
Cc Eric Dumazet <[email protected]>Cc Florian Westphal <[email protected]>Cc Andrew Vagin <[email protected]>
Quoted 2 lines
Reported-by Andrew Vagin <[email protected]>Signed-off-by Pablo Neira Ayuso <[email protected]>