diff options
author | Florian Westphal <fw@strlen.de> | 2017-04-14 20:31:08 +0200 |
---|---|---|
committer | Pablo Neira Ayuso <pablo@netfilter.org> | 2017-04-15 11:47:57 +0200 |
commit | cc41c84b7e7f2d7f6698bccc84890943fd021265 (patch) | |
tree | 08c09a05f63c6d146e0baf3f8f5f4752092c7622 /net/netfilter | |
parent | 6e354a5e565110c3a0eb7da3788340d4809a42b6 (diff) | |
download | op-kernel-dev-cc41c84b7e7f2d7f6698bccc84890943fd021265.zip op-kernel-dev-cc41c84b7e7f2d7f6698bccc84890943fd021265.tar.gz |
netfilter: kill the fake untracked conntrack objects
resurrect an old patch from Pablo Neira to remove the untracked objects.
Currently, there are four possible states of an skb wrt. conntrack.
1. No conntrack attached, ct is NULL.
2. Normal (kmem cache allocated) ct attached.
3. a template (kmalloc'd), not in any hash tables at any point in time
4. the 'untracked' conntrack, a percpu nf_conn object, tagged via
IPS_UNTRACKED_BIT in ct->status.
Untracked is supposed to be identical to case 1. It exists only
so users can check
-m conntrack --ctstate UNTRACKED vs.
-m conntrack --ctstate INVALID
e.g. attempts to set connmark on INVALID or UNTRACKED conntracks is
supposed to be a no-op.
Thus currently we need to check
ct == NULL || nf_ct_is_untracked(ct)
in a lot of places in order to avoid altering untracked objects.
The other consequence of the percpu untracked object is that all
-j NOTRACK (and, later, kfree_skb of such skbs) result in an atomic op
(inc/dec the untracked conntracks refcount).
This adds a new kernel-private ctinfo state, IP_CT_UNTRACKED, to
make the distinction instead.
The (few) places that care about packet invalid (ct is NULL) vs.
packet untracked now need to test ct == NULL vs. ctinfo == IP_CT_UNTRACKED,
but all other places can omit the nf_ct_is_untracked() check.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Diffstat (limited to 'net/netfilter')
-rw-r--r-- | net/netfilter/nf_conntrack_core.c | 48 | ||||
-rw-r--r-- | net/netfilter/nf_nat_core.c | 3 | ||||
-rw-r--r-- | net/netfilter/nft_ct.c | 14 | ||||
-rw-r--r-- | net/netfilter/xt_CT.c | 16 | ||||
-rw-r--r-- | net/netfilter/xt_conntrack.c | 11 | ||||
-rw-r--r-- | net/netfilter/xt_state.c | 13 |
6 files changed, 30 insertions, 75 deletions
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index bcf1d2a..03150f6 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -180,14 +180,6 @@ EXPORT_SYMBOL_GPL(nf_conntrack_htable_size); unsigned int nf_conntrack_max __read_mostly; seqcount_t nf_conntrack_generation __read_mostly; - -/* nf_conn must be 8 bytes aligned, as the 3 LSB bits are used - * for the nfctinfo. We cheat by (ab)using the PER CPU cache line - * alignment to enforce this. - */ -DEFINE_PER_CPU_ALIGNED(struct nf_conn, nf_conntrack_untracked); -EXPORT_PER_CPU_SYMBOL(nf_conntrack_untracked); - static unsigned int nf_conntrack_hash_rnd __read_mostly; static u32 hash_conntrack_raw(const struct nf_conntrack_tuple *tuple, @@ -1314,9 +1306,10 @@ nf_conntrack_in(struct net *net, u_int8_t pf, unsigned int hooknum, int ret; tmpl = nf_ct_get(skb, &ctinfo); - if (tmpl) { + if (tmpl || ctinfo == IP_CT_UNTRACKED) { /* Previously seen (loopback or untracked)? Ignore. */ - if (!nf_ct_is_template(tmpl)) { + if ((tmpl && !nf_ct_is_template(tmpl)) || + ctinfo == IP_CT_UNTRACKED) { NF_CT_STAT_INC_ATOMIC(net, ignore); return NF_ACCEPT; } @@ -1629,18 +1622,6 @@ void nf_ct_free_hashtable(void *hash, unsigned int size) } EXPORT_SYMBOL_GPL(nf_ct_free_hashtable); -static int untrack_refs(void) -{ - int cnt = 0, cpu; - - for_each_possible_cpu(cpu) { - struct nf_conn *ct = &per_cpu(nf_conntrack_untracked, cpu); - - cnt += atomic_read(&ct->ct_general.use) - 1; - } - return cnt; -} - void nf_conntrack_cleanup_start(void) { conntrack_gc_work.exiting = true; @@ -1650,8 +1631,6 @@ void nf_conntrack_cleanup_start(void) void nf_conntrack_cleanup_end(void) { RCU_INIT_POINTER(nf_ct_destroy, NULL); - while (untrack_refs() > 0) - schedule(); cancel_delayed_work_sync(&conntrack_gc_work.dwork); nf_ct_free_hashtable(nf_conntrack_hash, nf_conntrack_htable_size); @@ -1825,20 +1804,11 @@ EXPORT_SYMBOL_GPL(nf_conntrack_set_hashsize); module_param_call(hashsize, nf_conntrack_set_hashsize, param_get_uint, &nf_conntrack_htable_size, 0600); -void nf_ct_untracked_status_or(unsigned long bits) -{ - int cpu; - - for_each_possible_cpu(cpu) - per_cpu(nf_conntrack_untracked, cpu).status |= bits; -} -EXPORT_SYMBOL_GPL(nf_ct_untracked_status_or); - int nf_conntrack_init_start(void) { int max_factor = 8; int ret = -ENOMEM; - int i, cpu; + int i; seqcount_init(&nf_conntrack_generation); @@ -1921,15 +1891,6 @@ int nf_conntrack_init_start(void) if (ret < 0) goto err_proto; - /* Set up fake conntrack: to never be deleted, not in any hashes */ - for_each_possible_cpu(cpu) { - struct nf_conn *ct = &per_cpu(nf_conntrack_untracked, cpu); - write_pnet(&ct->ct_net, &init_net); - atomic_set(&ct->ct_general.use, 1); - } - /* - and look it like as a confirmed connection */ - nf_ct_untracked_status_or(IPS_CONFIRMED | IPS_UNTRACKED); - conntrack_gc_work_init(&conntrack_gc_work); queue_delayed_work(system_long_wq, &conntrack_gc_work.dwork, HZ); @@ -1977,6 +1938,7 @@ int nf_conntrack_init_net(struct net *net) int ret = -ENOMEM; int cpu; + BUILD_BUG_ON(IP_CT_UNTRACKED == IP_CT_NUMBER); atomic_set(&net->ct.count, 0); net->ct.pcpu_lists = alloc_percpu(struct ct_pcpu); diff --git a/net/netfilter/nf_nat_core.c b/net/netfilter/nf_nat_core.c index 5e35643..9cbf49f 100644 --- a/net/netfilter/nf_nat_core.c +++ b/net/netfilter/nf_nat_core.c @@ -861,9 +861,6 @@ static int __init nf_nat_init(void) nf_ct_helper_expectfn_register(&follow_master_nat); - /* Initialize fake conntrack so that NAT will skip it */ - nf_ct_untracked_status_or(IPS_NAT_DONE_MASK); - BUG_ON(nfnetlink_parse_nat_setup_hook != NULL); RCU_INIT_POINTER(nfnetlink_parse_nat_setup_hook, nfnetlink_parse_nat_setup); diff --git a/net/netfilter/nft_ct.c b/net/netfilter/nft_ct.c index 6e23dbb..6c6fd48 100644 --- a/net/netfilter/nft_ct.c +++ b/net/netfilter/nft_ct.c @@ -72,12 +72,12 @@ static void nft_ct_get_eval(const struct nft_expr *expr, switch (priv->key) { case NFT_CT_STATE: - if (ct == NULL) - state = NF_CT_STATE_INVALID_BIT; - else if (nf_ct_is_untracked(ct)) + if (ct) + state = NF_CT_STATE_BIT(ctinfo); + else if (ctinfo == IP_CT_UNTRACKED) state = NF_CT_STATE_UNTRACKED_BIT; else - state = NF_CT_STATE_BIT(ctinfo); + state = NF_CT_STATE_INVALID_BIT; *dest = state; return; default: @@ -718,12 +718,10 @@ static void nft_notrack_eval(const struct nft_expr *expr, ct = nf_ct_get(pkt->skb, &ctinfo); /* Previously seen (loopback or untracked)? Ignore. */ - if (ct) + if (ct || ctinfo == IP_CT_UNTRACKED) return; - ct = nf_ct_untracked_get(); - atomic_inc(&ct->ct_general.use); - nf_ct_set(skb, ct, IP_CT_NEW); + nf_ct_set(skb, ct, IP_CT_UNTRACKED); } static struct nft_expr_type nft_notrack_type; diff --git a/net/netfilter/xt_CT.c b/net/netfilter/xt_CT.c index b008db0..3cbe1bc 100644 --- a/net/netfilter/xt_CT.c +++ b/net/netfilter/xt_CT.c @@ -26,11 +26,12 @@ static inline int xt_ct_target(struct sk_buff *skb, struct nf_conn *ct) if (skb->_nfct != 0) return XT_CONTINUE; - /* special case the untracked ct : we want the percpu object */ - if (!ct) - ct = nf_ct_untracked_get(); - atomic_inc(&ct->ct_general.use); - nf_ct_set(skb, ct, IP_CT_NEW); + if (ct) { + atomic_inc(&ct->ct_general.use); + nf_ct_set(skb, ct, IP_CT_NEW); + } else { + nf_ct_set(skb, ct, IP_CT_UNTRACKED); + } return XT_CONTINUE; } @@ -335,7 +336,7 @@ static void xt_ct_tg_destroy(const struct xt_tgdtor_param *par, struct nf_conn *ct = info->ct; struct nf_conn_help *help; - if (ct && !nf_ct_is_untracked(ct)) { + if (ct) { help = nfct_help(ct); if (help) module_put(help->helper->me); @@ -412,8 +413,7 @@ notrack_tg(struct sk_buff *skb, const struct xt_action_param *par) if (skb->_nfct != 0) return XT_CONTINUE; - nf_ct_set(skb, nf_ct_untracked_get(), IP_CT_NEW); - nf_conntrack_get(skb_nfct(skb)); + nf_ct_set(skb, NULL, IP_CT_UNTRACKED); return XT_CONTINUE; } diff --git a/net/netfilter/xt_conntrack.c b/net/netfilter/xt_conntrack.c index c0fb217..39cf1d0 100644 --- a/net/netfilter/xt_conntrack.c +++ b/net/netfilter/xt_conntrack.c @@ -172,12 +172,11 @@ conntrack_mt(const struct sk_buff *skb, struct xt_action_param *par, ct = nf_ct_get(skb, &ctinfo); - if (ct) { - if (nf_ct_is_untracked(ct)) - statebit = XT_CONNTRACK_STATE_UNTRACKED; - else - statebit = XT_CONNTRACK_STATE_BIT(ctinfo); - } else + if (ct) + statebit = XT_CONNTRACK_STATE_BIT(ctinfo); + else if (ctinfo == IP_CT_UNTRACKED) + statebit = XT_CONNTRACK_STATE_UNTRACKED; + else statebit = XT_CONNTRACK_STATE_INVALID; if (info->match_flags & XT_CONNTRACK_STATE) { diff --git a/net/netfilter/xt_state.c b/net/netfilter/xt_state.c index 5746a33..5fbd791 100644 --- a/net/netfilter/xt_state.c +++ b/net/netfilter/xt_state.c @@ -28,14 +28,13 @@ state_mt(const struct sk_buff *skb, struct xt_action_param *par) unsigned int statebit; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); - if (!ct) + if (ct) + statebit = XT_STATE_BIT(ctinfo); + else if (ctinfo == IP_CT_UNTRACKED) + statebit = XT_STATE_UNTRACKED; + else statebit = XT_STATE_INVALID; - else { - if (nf_ct_is_untracked(ct)) - statebit = XT_STATE_UNTRACKED; - else - statebit = XT_STATE_BIT(ctinfo); - } + return (sinfo->statemask & statebit); } |