summaryrefslogtreecommitdiffstats
path: root/net/ipv4
diff options
context:
space:
mode:
Diffstat (limited to 'net/ipv4')
-rw-r--r--net/ipv4/Makefile2
-rw-r--r--net/ipv4/af_inet.c2
-rw-r--r--net/ipv4/fib_trie.c1
-rw-r--r--net/ipv4/igmp.c47
-rw-r--r--net/ipv4/inet_fragment.c141
-rw-r--r--net/ipv4/ip_fragment.c67
-rw-r--r--net/ipv4/ip_output.c313
-rw-r--r--net/ipv4/ip_sockglue.c2
-rw-r--r--net/ipv4/netfilter/nft_fib_ipv4.c23
-rw-r--r--net/ipv4/nexthop.c1479
-rw-r--r--net/ipv4/proc.c5
-rw-r--r--net/ipv4/sysctl_net_ipv4.c93
-rw-r--r--net/ipv4/tcp.c29
-rw-r--r--net/ipv4/tcp_fastopen.c233
-rw-r--r--net/ipv4/tcp_input.c2
15 files changed, 2080 insertions, 359 deletions
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index 000a61994c8f..d57ecfaf89d4 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -14,7 +14,7 @@ obj-y := route.o inetpeer.o protocol.o \
udp_offload.o arp.o icmp.o devinet.o af_inet.o igmp.o \
fib_frontend.o fib_semantics.o fib_trie.o fib_notifier.o \
inet_fragment.o ping.o ip_tunnel_core.o gre_offload.o \
- metrics.o netlink.o
+ metrics.o netlink.o nexthop.o
obj-$(CONFIG_BPFILTER) += bpfilter/
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 5183a2daba64..aff93e7cdb31 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -428,8 +428,8 @@ int inet_release(struct socket *sock)
if (sock_flag(sk, SOCK_LINGER) &&
!(current->flags & PF_EXITING))
timeout = sk->sk_lingertime;
- sock->sk = NULL;
sk->sk_prot->close(sk, timeout);
+ sock->sk = NULL;
}
return 0;
}
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index ea7df7ebf597..b53ecef89d59 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1961,7 +1961,6 @@ static void __fib_info_notify_update(struct net *net, struct fib_table *tb,
if (IS_TRIE(pn))
break;
- n = pn;
pn = node_parent(pn);
cindex = get_index(pkey, pn);
continue;
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 6c2febc39dca..eb03153dfe12 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -188,6 +188,17 @@ static void ip_ma_put(struct ip_mc_list *im)
pmc != NULL; \
pmc = rtnl_dereference(pmc->next_rcu))
+static void ip_sf_list_clear_all(struct ip_sf_list *psf)
+{
+ struct ip_sf_list *next;
+
+ while (psf) {
+ next = psf->sf_next;
+ kfree(psf);
+ psf = next;
+ }
+}
+
#ifdef CONFIG_IP_MULTICAST
/*
@@ -633,6 +644,13 @@ static void igmpv3_clear_zeros(struct ip_sf_list **ppsf)
}
}
+static void kfree_pmc(struct ip_mc_list *pmc)
+{
+ ip_sf_list_clear_all(pmc->sources);
+ ip_sf_list_clear_all(pmc->tomb);
+ kfree(pmc);
+}
+
static void igmpv3_send_cr(struct in_device *in_dev)
{
struct ip_mc_list *pmc, *pmc_prev, *pmc_next;
@@ -669,7 +687,7 @@ static void igmpv3_send_cr(struct in_device *in_dev)
else
in_dev->mc_tomb = pmc_next;
in_dev_put(pmc->interface);
- kfree(pmc);
+ kfree_pmc(pmc);
} else
pmc_prev = pmc;
}
@@ -1215,14 +1233,18 @@ static void igmpv3_del_delrec(struct in_device *in_dev, struct ip_mc_list *im)
im->interface = pmc->interface;
if (im->sfmode == MCAST_INCLUDE) {
im->tomb = pmc->tomb;
+ pmc->tomb = NULL;
+
im->sources = pmc->sources;
+ pmc->sources = NULL;
+
for (psf = im->sources; psf; psf = psf->sf_next)
psf->sf_crcount = in_dev->mr_qrv ?: net->ipv4.sysctl_igmp_qrv;
} else {
im->crcount = in_dev->mr_qrv ?: net->ipv4.sysctl_igmp_qrv;
}
in_dev_put(pmc->interface);
- kfree(pmc);
+ kfree_pmc(pmc);
}
spin_unlock_bh(&im->lock);
}
@@ -1243,21 +1265,18 @@ static void igmpv3_clear_delrec(struct in_device *in_dev)
nextpmc = pmc->next;
ip_mc_clear_src(pmc);
in_dev_put(pmc->interface);
- kfree(pmc);
+ kfree_pmc(pmc);
}
/* clear dead sources, too */
rcu_read_lock();
for_each_pmc_rcu(in_dev, pmc) {
- struct ip_sf_list *psf, *psf_next;
+ struct ip_sf_list *psf;
spin_lock_bh(&pmc->lock);
psf = pmc->tomb;
pmc->tomb = NULL;
spin_unlock_bh(&pmc->lock);
- for (; psf; psf = psf_next) {
- psf_next = psf->sf_next;
- kfree(psf);
- }
+ ip_sf_list_clear_all(psf);
}
rcu_read_unlock();
}
@@ -2123,7 +2142,7 @@ static int ip_mc_add_src(struct in_device *in_dev, __be32 *pmca, int sfmode,
static void ip_mc_clear_src(struct ip_mc_list *pmc)
{
- struct ip_sf_list *psf, *nextpsf, *tomb, *sources;
+ struct ip_sf_list *tomb, *sources;
spin_lock_bh(&pmc->lock);
tomb = pmc->tomb;
@@ -2135,14 +2154,8 @@ static void ip_mc_clear_src(struct ip_mc_list *pmc)
pmc->sfcount[MCAST_EXCLUDE] = 1;
spin_unlock_bh(&pmc->lock);
- for (psf = tomb; psf; psf = nextpsf) {
- nextpsf = psf->sf_next;
- kfree(psf);
- }
- for (psf = sources; psf; psf = nextpsf) {
- nextpsf = psf->sf_next;
- kfree(psf);
- }
+ ip_sf_list_clear_all(tomb);
+ ip_sf_list_clear_all(sources);
}
/* Join a multicast group
diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c
index 737808e27f8b..35e9784fab4e 100644
--- a/net/ipv4/inet_fragment.c
+++ b/net/ipv4/inet_fragment.c
@@ -110,48 +110,101 @@ int inet_frags_init(struct inet_frags *f)
if (!f->frags_cachep)
return -ENOMEM;
+ refcount_set(&f->refcnt, 1);
+ init_completion(&f->completion);
return 0;
}
EXPORT_SYMBOL(inet_frags_init);
void inet_frags_fini(struct inet_frags *f)
{
- /* We must wait that all inet_frag_destroy_rcu() have completed. */
- rcu_barrier();
+ if (refcount_dec_and_test(&f->refcnt))
+ complete(&f->completion);
+
+ wait_for_completion(&f->completion);
kmem_cache_destroy(f->frags_cachep);
f->frags_cachep = NULL;
}
EXPORT_SYMBOL(inet_frags_fini);
+/* called from rhashtable_free_and_destroy() at netns_frags dismantle */
static void inet_frags_free_cb(void *ptr, void *arg)
{
struct inet_frag_queue *fq = ptr;
+ int count;
- /* If we can not cancel the timer, it means this frag_queue
- * is already disappearing, we have nothing to do.
- * Otherwise, we own a refcount until the end of this function.
- */
- if (!del_timer(&fq->timer))
- return;
+ count = del_timer_sync(&fq->timer) ? 1 : 0;
spin_lock_bh(&fq->lock);
if (!(fq->flags & INET_FRAG_COMPLETE)) {
fq->flags |= INET_FRAG_COMPLETE;
- refcount_dec(&fq->refcnt);
+ count++;
+ } else if (fq->flags & INET_FRAG_HASH_DEAD) {
+ count++;
}
spin_unlock_bh(&fq->lock);
- inet_frag_put(fq);
+ if (refcount_sub_and_test(count, &fq->refcnt))
+ inet_frag_destroy(fq);
}
-void inet_frags_exit_net(struct netns_frags *nf)
+static void fqdir_rwork_fn(struct work_struct *work)
{
- nf->high_thresh = 0; /* prevent creation of new frags */
+ struct fqdir *fqdir = container_of(to_rcu_work(work),
+ struct fqdir, destroy_rwork);
+ struct inet_frags *f = fqdir->f;
+
+ rhashtable_free_and_destroy(&fqdir->rhashtable, inet_frags_free_cb, NULL);
+
+ /* We need to make sure all ongoing call_rcu(..., inet_frag_destroy_rcu)
+ * have completed, since they need to dereference fqdir.
+ * Would it not be nice to have kfree_rcu_barrier() ? :)
+ */
+ rcu_barrier();
+
+ if (refcount_dec_and_test(&f->refcnt))
+ complete(&f->completion);
- rhashtable_free_and_destroy(&nf->rhashtable, inet_frags_free_cb, NULL);
+ kfree(fqdir);
}
-EXPORT_SYMBOL(inet_frags_exit_net);
+
+int fqdir_init(struct fqdir **fqdirp, struct inet_frags *f, struct net *net)
+{
+ struct fqdir *fqdir = kzalloc(sizeof(*fqdir), GFP_KERNEL);
+ int res;
+
+ if (!fqdir)
+ return -ENOMEM;
+ fqdir->f = f;
+ fqdir->net = net;
+ res = rhashtable_init(&fqdir->rhashtable, &fqdir->f->rhash_params);
+ if (res < 0) {
+ kfree(fqdir);
+ return res;
+ }
+ refcount_inc(&f->refcnt);
+ *fqdirp = fqdir;
+ return 0;
+}
+EXPORT_SYMBOL(fqdir_init);
+
+void fqdir_exit(struct fqdir *fqdir)
+{
+ fqdir->high_thresh = 0; /* prevent creation of new frags */
+
+ fqdir->dead = true;
+
+ /* call_rcu is supposed to provide memory barrier semantics,
+ * separating the setting of fqdir->dead with the destruction
+ * work. This implicit barrier is paired with inet_frag_kill().
+ */
+
+ INIT_RCU_WORK(&fqdir->destroy_rwork, fqdir_rwork_fn);
+ queue_rcu_work(system_wq, &fqdir->destroy_rwork);
+
+}
+EXPORT_SYMBOL(fqdir_exit);
void inet_frag_kill(struct inet_frag_queue *fq)
{
@@ -159,11 +212,23 @@ void inet_frag_kill(struct inet_frag_queue *fq)
refcount_dec(&fq->refcnt);
if (!(fq->flags & INET_FRAG_COMPLETE)) {
- struct netns_frags *nf = fq->net;
+ struct fqdir *fqdir = fq->fqdir;
fq->flags |= INET_FRAG_COMPLETE;
- rhashtable_remove_fast(&nf->rhashtable, &fq->node, nf->f->rhash_params);
- refcount_dec(&fq->refcnt);
+ rcu_read_lock();
+ /* The RCU read lock provides a memory barrier
+ * guaranteeing that if fqdir->dead is false then
+ * the hash table destruction will not start until
+ * after we unlock. Paired with inet_frags_exit_net().
+ */
+ if (!fqdir->dead) {
+ rhashtable_remove_fast(&fqdir->rhashtable, &fq->node,
+ fqdir->f->rhash_params);
+ refcount_dec(&fq->refcnt);
+ } else {
+ fq->flags |= INET_FRAG_HASH_DEAD;
+ }
+ rcu_read_unlock();
}
}
EXPORT_SYMBOL(inet_frag_kill);
@@ -172,7 +237,7 @@ static void inet_frag_destroy_rcu(struct rcu_head *head)
{
struct inet_frag_queue *q = container_of(head, struct inet_frag_queue,
rcu);
- struct inet_frags *f = q->net->f;
+ struct inet_frags *f = q->fqdir->f;
if (f->destructor)
f->destructor(q);
@@ -203,7 +268,7 @@ EXPORT_SYMBOL(inet_frag_rbtree_purge);
void inet_frag_destroy(struct inet_frag_queue *q)
{
- struct netns_frags *nf;
+ struct fqdir *fqdir;
unsigned int sum, sum_truesize = 0;
struct inet_frags *f;
@@ -211,18 +276,18 @@ void inet_frag_destroy(struct inet_frag_queue *q)
WARN_ON(del_timer(&q->timer) != 0);
/* Release all fragment data. */
- nf = q->net;
- f = nf->f;
+ fqdir = q->fqdir;
+ f = fqdir->f;
sum_truesize = inet_frag_rbtree_purge(&q->rb_fragments);
sum = sum_truesize + f->qsize;
call_rcu(&q->rcu, inet_frag_destroy_rcu);
- sub_frag_mem_limit(nf, sum);
+ sub_frag_mem_limit(fqdir, sum);
}
EXPORT_SYMBOL(inet_frag_destroy);
-static struct inet_frag_queue *inet_frag_alloc(struct netns_frags *nf,
+static struct inet_frag_queue *inet_frag_alloc(struct fqdir *fqdir,
struct inet_frags *f,
void *arg)
{
@@ -232,9 +297,9 @@ static struct inet_frag_queue *inet_frag_alloc(struct netns_frags *nf,
if (!q)
return NULL;
- q->net = nf;
+ q->fqdir = fqdir;
f->constructor(q, arg);
- add_frag_mem_limit(nf, f->qsize);
+ add_frag_mem_limit(fqdir, f->qsize);
timer_setup(&q->timer, f->frag_expire, 0);
spin_lock_init(&q->lock);
@@ -243,21 +308,21 @@ static struct inet_frag_queue *inet_frag_alloc(struct netns_frags *nf,
return q;
}
-static struct inet_frag_queue *inet_frag_create(struct netns_frags *nf,
+static struct inet_frag_queue *inet_frag_create(struct fqdir *fqdir,
void *arg,
struct inet_frag_queue **prev)
{
- struct inet_frags *f = nf->f;
+ struct inet_frags *f = fqdir->f;
struct inet_frag_queue *q;
- q = inet_frag_alloc(nf, f, arg);
+ q = inet_frag_alloc(fqdir, f, arg);
if (!q) {
*prev = ERR_PTR(-ENOMEM);
return NULL;
}
- mod_timer(&q->timer, jiffies + nf->timeout);
+ mod_timer(&q->timer, jiffies + fqdir->timeout);
- *prev = rhashtable_lookup_get_insert_key(&nf->rhashtable, &q->key,
+ *prev = rhashtable_lookup_get_insert_key(&fqdir->rhashtable, &q->key,
&q->node, f->rhash_params);
if (*prev) {
q->flags |= INET_FRAG_COMPLETE;
@@ -269,18 +334,18 @@ static struct inet_frag_queue *inet_frag_create(struct netns_frags *nf,
}
/* TODO : call from rcu_read_lock() and no longer use refcount_inc_not_zero() */
-struct inet_frag_queue *inet_frag_find(struct netns_frags *nf, void *key)
+struct inet_frag_queue *inet_frag_find(struct fqdir *fqdir, void *key)
{
struct inet_frag_queue *fq = NULL, *prev;
- if (!nf->high_thresh || frag_mem_limit(nf) > nf->high_thresh)
+ if (!fqdir->high_thresh || frag_mem_limit(fqdir) > fqdir->high_thresh)
return NULL;
rcu_read_lock();
- prev = rhashtable_lookup(&nf->rhashtable, key, nf->f->rhash_params);
+ prev = rhashtable_lookup(&fqdir->rhashtable, key, fqdir->f->rhash_params);
if (!prev)
- fq = inet_frag_create(nf, key, &prev);
+ fq = inet_frag_create(fqdir, key, &prev);
if (prev && !IS_ERR(prev)) {
fq = prev;
if (!refcount_inc_not_zero(&fq->refcnt))
@@ -391,7 +456,7 @@ void *inet_frag_reasm_prepare(struct inet_frag_queue *q, struct sk_buff *skb,
delta += head->truesize;
if (delta)
- add_frag_mem_limit(q->net, delta);
+ add_frag_mem_limit(q->fqdir, delta);
/* If the first fragment is fragmented itself, we split
* it to two chunks: the first with data and paged part
@@ -413,7 +478,7 @@ void *inet_frag_reasm_prepare(struct inet_frag_queue *q, struct sk_buff *skb,
head->truesize += clone->truesize;
clone->csum = 0;
clone->ip_summed = head->ip_summed;
- add_frag_mem_limit(q->net, clone->truesize);
+ add_frag_mem_limit(q->fqdir, clone->truesize);
skb_shinfo(head)->frag_list = clone;
nextp = &clone->next;
} else {
@@ -466,7 +531,7 @@ void inet_frag_reasm_finish(struct inet_frag_queue *q, struct sk_buff *head,
rbn = rbnext;
}
}
- sub_frag_mem_limit(q->net, head->truesize);
+ sub_frag_mem_limit(q->fqdir, head->truesize);
*nextp = NULL;
skb_mark_not_on_list(head);
@@ -494,7 +559,7 @@ struct sk_buff *inet_frag_pull_head(struct inet_frag_queue *q)
if (head == q->fragments_tail)
q->fragments_tail = NULL;
- sub_frag_mem_limit(q->net, head->truesize);
+ sub_frag_mem_limit(q->fqdir, head->truesize);
return head;
}
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index cf2b0a6a3337..1ffaec056821 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -82,15 +82,13 @@ static int ip_frag_reasm(struct ipq *qp, struct sk_buff *skb,
static void ip4_frag_init(struct inet_frag_queue *q, const void *a)
{
struct ipq *qp = container_of(q, struct ipq, q);
- struct netns_ipv4 *ipv4 = container_of(q->net, struct netns_ipv4,
- frags);
- struct net *net = container_of(ipv4, struct net, ipv4);
+ struct net *net = q->fqdir->net;
const struct frag_v4_compare_key *key = a;
q->key.v4 = *key;
qp->ecn = 0;
- qp->peer = q->net->max_dist ?
+ qp->peer = q->fqdir->max_dist ?
inet_getpeer_v4(net->ipv4.peers, key->saddr, key->vif, 1) :
NULL;
}
@@ -142,7 +140,7 @@ static void ip_expire(struct timer_list *t)
int err;
qp = container_of(frag, struct ipq, q);
- net = container_of(qp->q.net, struct net, ipv4.frags);
+ net = qp->q.fqdir->net;
rcu_read_lock();
spin_lock(&qp->q.lock);
@@ -211,7 +209,7 @@ static struct ipq *ip_find(struct net *net, struct iphdr *iph,
};
struct inet_frag_queue *q;
- q = inet_frag_find(&net->ipv4.frags, &key);
+ q = inet_frag_find(net->ipv4.fqdir, &key);
if (!q)
return NULL;
@@ -222,7 +220,7 @@ static struct ipq *ip_find(struct net *net, struct iphdr *iph,
static int ip_frag_too_far(struct ipq *qp)
{
struct inet_peer *peer = qp->peer;
- unsigned int max = qp->q.net->max_dist;
+ unsigned int max = qp->q.fqdir->max_dist;
unsigned int start, end;
int rc;
@@ -236,12 +234,8 @@ static int ip_frag_too_far(struct ipq *qp)
rc = qp->q.fragments_tail && (end - start) > max;
- if (rc) {
- struct net *net;
-
- net = container_of(qp->q.net, struct net, ipv4.frags);
- __IP_INC_STATS(net, IPSTATS_MIB_REASMFAILS);
- }
+ if (rc)
+ __IP_INC_STATS(qp->q.fqdir->net, IPSTATS_MIB_REASMFAILS);
return rc;
}
@@ -250,13 +244,13 @@ static int ip_frag_reinit(struct ipq *qp)
{
unsigned int sum_truesize = 0;
- if (!mod_timer(&qp->q.timer, jiffies + qp->q.net->timeout)) {
+ if (!mod_timer(&qp->q.timer, jiffies + qp->q.fqdir->timeout)) {
refcount_inc(&qp->q.refcnt);
return -ETIMEDOUT;
}
sum_truesize = inet_frag_rbtree_purge(&qp->q.rb_fragments);
- sub_frag_mem_limit(qp->q.net, sum_truesize);
+ sub_frag_mem_limit(qp->q.fqdir, sum_truesize);
qp->q.flags = 0;
qp->q.len = 0;
@@ -273,7 +267,7 @@ static int ip_frag_reinit(struct ipq *qp)
/* Add new segment to existing queue. */
static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb)
{
- struct net *net = container_of(qp->q.net, struct net, ipv4.frags);
+ struct net *net = qp->q.fqdir->net;
int ihl, end, flags, offset;
struct sk_buff *prev_tail;
struct net_device *dev;
@@ -352,7 +346,7 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb)
qp->q.stamp = skb->tstamp;
qp->q.meat += skb->len;
qp->ecn |= ecn;
- add_frag_mem_limit(qp->q.net, skb->truesize);
+ add_frag_mem_limit(qp->q.fqdir, skb->truesize);
if (offset == 0)
qp->q.flags |= INET_FRAG_FIRST_IN;
@@ -399,7 +393,7 @@ err:
static int ip_frag_reasm(struct ipq *qp, struct sk_buff *skb,
struct sk_buff *prev_tail, struct net_device *dev)
{
- struct net *net = container_of(qp->q.net, struct net, ipv4.frags);
+ struct net *net = qp->q.fqdir->net;
struct iphdr *iph;
void *reasm_data;
int len, err;
@@ -544,30 +538,24 @@ static int dist_min;
static struct ctl_table ip4_frags_ns_ctl_table[] = {
{
.procname = "ipfrag_high_thresh",
- .data = &init_net.ipv4.frags.high_thresh,
.maxlen = sizeof(unsigned long),
.mode = 0644,
.proc_handler = proc_doulongvec_minmax,
- .extra1 = &init_net.ipv4.frags.low_thresh
},
{
.procname = "ipfrag_low_thresh",
- .data = &init_net.ipv4.frags.low_thresh,
.maxlen = sizeof(unsigned long),
.mode = 0644,
.proc_handler = proc_doulongvec_minmax,
- .extra2 = &init_net.ipv4.frags.high_thresh
},
{
.procname = "ipfrag_time",
- .data = &init_net.ipv4.frags.timeout,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_jiffies,
},
{
.procname = "ipfrag_max_dist",
- .data = &init_net.ipv4.frags.max_dist,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_minmax,
@@ -600,13 +588,13 @@ static int __net_init ip4_frags_ns_ctl_register(struct net *net)
if (!table)
goto err_alloc;
- table[0].data = &net->ipv4.frags.high_thresh;
- table[0].extra1 = &net->ipv4.frags.low_thresh;
- table[1].data = &net->ipv4.frags.low_thresh;
- table[1].extra2 = &net->ipv4.frags.high_thresh;
- table[2].data = &net->ipv4.frags.timeout;
- table[3].data = &net->ipv4.frags.max_dist;
}
+ table[0].data = &net->ipv4.fqdir->high_thresh;
+ table[0].extra1 = &net->ipv4.fqdir->low_thresh;
+ table[1].data = &net->ipv4.fqdir->low_thresh;
+ table[1].extra2 = &net->ipv4.fqdir->high_thresh;
+ table[2].data = &net->ipv4.fqdir->timeout;
+ table[3].data = &net->ipv4.fqdir->max_dist;
hdr = register_net_sysctl(net, "net/ipv4", table);
if (!hdr)
@@ -654,6 +642,9 @@ static int __net_init ipv4_frags_init_net(struct net *net)
{
int res;
+ res = fqdir_init(&net->ipv4.fqdir, &ip4_frags, net);
+ if (res < 0)
+ return res;
/* Fragment cache limits.
*
* The fragment memory accounting code, (tries to) account for
@@ -668,31 +659,27 @@ static int __net_init ipv4_frags_init_net(struct net *net)
* we will prune down to 3MB, making room for approx 8 big 64K
* fragments 8x128k.
*/
- net->ipv4.frags.high_thresh = 4 * 1024 * 1024;
- net->ipv4.frags.low_thresh = 3 * 1024 * 1024;
+ net->ipv4.fqdir->high_thresh = 4 * 1024 * 1024;
+ net->ipv4.fqdir->low_thresh = 3 * 1024 * 1024;
/*
* Important NOTE! Fragment queue must be destroyed before MSL expires.
* RFC791 is wrong proposing to prolongate timer each fragment arrival
* by TTL.
*/
- net->ipv4.frags.timeout = IP_FRAG_TIME;
+ net->ipv4.fqdir->timeout = IP_FRAG_TIME;
- net->ipv4.frags.max_dist = 64;
- net->ipv4.frags.f = &ip4_frags;
+ net->ipv4.fqdir->max_dist = 64;
- res = inet_frags_init_net(&net->ipv4.frags);
- if (res < 0)
- return res;
res = ip4_frags_ns_ctl_register(net);
if (res < 0)
- inet_frags_exit_net(&net->ipv4.frags);
+ fqdir_exit(net->ipv4.fqdir);
return res;
}
static void __net_exit ipv4_frags_exit_net(struct net *net)
{
ip4_frags_ns_ctl_unregister(net);
- inet_frags_exit_net(&net->ipv4.frags);
+ fqdir_exit(net->ipv4.fqdir);
}
static struct pernet_operations ip4_frags_ops = {
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 1217a53381c2..ceca5285d9b4 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -537,9 +537,6 @@ static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from)
skb_copy_hash(to, from);
- /* Copy the flags to each fragment. */
- IPCB(to)->flags = IPCB(from)->flags;
-
#ifdef CONFIG_NET_SCHED
to->tc_index = from->tc_index;
#endif
@@ -573,6 +570,176 @@ static int ip_fragment(struct net *net, struct sock *sk, struct sk_buff *skb,
return ip_do_fragment(net, sk, skb, output);
}
+void ip_fraglist_init(struct sk_buff *skb, struct iphdr *iph,
+ unsigned int hlen, struct ip_fraglist_iter *iter)
+{
+ unsigned int first_len = skb_pagelen(skb);
+
+ iter->frag_list = skb_shinfo(skb)->frag_list;
+ iter->frag = iter->frag_list;
+ skb_frag_list_init(skb);
+
+ iter->offset = 0;
+ iter->iph = iph;
+ iter->hlen = hlen;
+
+ skb->data_len = first_len - skb_headlen(skb);
+ skb->len = first_len;
+ iph->tot_len = htons(first_len);
+ iph->frag_off = htons(IP_MF);
+ ip_send_check(iph);
+}
+EXPORT_SYMBOL(ip_fraglist_init);
+
+static void ip_fraglist_ipcb_prepare(struct sk_buff *skb,
+ struct ip_fraglist_iter *iter)
+{
+ struct sk_buff *to = iter->frag;
+
+ /* Copy the flags to each fragment. */
+ IPCB(to)->flags = IPCB(skb)->flags;
+
+ if (iter->offset == 0)
+ ip_options_fragment(to);
+}
+
+void ip_fraglist_prepare(struct sk_buff *skb, struct ip_fraglist_iter *iter)
+{
+ unsigned int hlen = iter->hlen;
+ struct iphdr *iph = iter->iph;
+ struct sk_buff *frag;
+
+ frag = iter->frag;
+ frag->ip_summed = CHECKSUM_NONE;
+ skb_reset_transport_header(frag);
+ __skb_push(frag, hlen);
+ skb_reset_network_header(frag);
+ memcpy(skb_network_header(frag), iph, hlen);
+ iter->iph = ip_hdr(frag);
+ iph = iter->iph;
+ iph->tot_len = htons(frag->len);
+ ip_copy_metadata(frag, skb);
+ iter->offset += skb->len - hlen;
+ iph->frag_off = htons(iter->offset >> 3);
+ if (frag->next)
+ iph->frag_off |= htons(IP_MF);
+ /* Ready, complete checksum */
+ ip_send_check(iph);
+}
+EXPORT_SYMBOL(ip_fraglist_prepare);
+
+void ip_frag_init(struct sk_buff *skb, unsigned int hlen,
+ unsigned int ll_rs, unsigned int mtu,
+ struct ip_frag_state *state)
+{
+ struct iphdr *iph = ip_hdr(skb);
+
+ state->hlen = hlen;
+ state->ll_rs = ll_rs;
+ state->mtu = mtu;
+
+ state->left = skb->len - hlen; /* Space per frame */
+ state->ptr = hlen; /* Where to start from */
+
+ state->offset = (ntohs(iph->frag_off) & IP_OFFSET) << 3;
+ state->not_last_frag = iph->frag_off & htons(IP_MF);
+}
+EXPORT_SYMBOL(ip_frag_init);
+
+static void ip_frag_ipcb(struct sk_buff *from, struct sk_buff *to,
+ bool first_frag, struct ip_frag_state *state)
+{
+ /* Copy the flags to each fragment. */
+ IPCB(to)->flags = IPCB(from)->flags;
+
+ if (IPCB(from)->flags & IPSKB_FRAG_PMTU)
+ state->iph->frag_off |= htons(IP_DF);
+
+ /* ANK: dirty, but effective trick. Upgrade options only if
+ * the segment to be fragmented was THE FIRST (otherwise,
+ * options are already fixed) and make it ONCE
+ * on the initial skb, so that all the following fragments
+ * will inherit fixed options.
+ */
+ if (first_frag)
+ ip_options_fragment(from);
+}
+
+struct sk_buff *ip_frag_next(struct sk_buff *skb, struct ip_frag_state *state)
+{
+ unsigned int len = state->left;
+ struct sk_buff *skb2;
+ struct iphdr *iph;
+
+ len = state->left;
+ /* IF: it doesn't fit, use 'mtu' - the data space left */
+ if (len > state->mtu)
+ len = state->mtu;
+ /* IF: we are not sending up to and including the packet end
+ then align the next start on an eight byte boundary */
+ if (len < state->left) {
+ len &= ~7;
+ }
+
+ /* Allocate buffer */
+ skb2 = alloc_skb(len + state->hlen + state->ll_rs, GFP_ATOMIC);
+ if (!skb2)
+ return ERR_PTR(-ENOMEM);
+
+ /*
+ * Set up data on packet
+ */
+
+ ip_copy_metadata(skb2, skb);
+ skb_reserve(skb2, state->ll_rs);
+ skb_put(skb2, len + state->hlen);
+ skb_reset_network_header(skb2);
+ skb2->transport_header = skb2->network_header + state->hlen;
+
+ /*
+ * Charge the memory for the fragment to any owner
+ * it might possess
+ */
+
+ if (skb->sk)
+ skb_set_owner_w(skb2, skb->sk);
+
+ /*
+ * Copy the packet header into the new buffer.
+ */
+
+ skb_copy_from_linear_data(skb, skb_network_header(skb2), state->hlen);
+
+ /*
+ * Copy a block of the IP datagram.
+ */
+ if (skb_copy_bits(skb, state->ptr, skb_transport_header(skb2), len))
+ BUG();
+ state->left -= len;
+
+ /*
+ * Fill in the new header fields.
+ */
+ iph = ip_hdr(skb2);
+ iph->frag_off = htons((state->offset >> 3));
+
+ /*
+ * Added AC : If we are fragmenting a fragment that's not the
+ * last fragment then keep MF on each bit
+ */
+ if (state->left > 0 || state->not_last_frag)
+ iph->frag_off |= htons(IP_MF);
+ state->ptr += len;
+ state->offset += len;
+
+ iph->tot_len = htons(len + state->hlen);
+
+ ip_send_check(iph);
+
+ return skb2;
+}
+EXPORT_SYMBOL(ip_frag_next);
+
/*
* This IP datagram is too large to be sent in one piece. Break it up into
* smaller pieces (each of size equal to IP header plus
@@ -584,12 +751,11 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb,
int (*output)(struct net *, struct sock *, struct sk_buff *))
{
struct iphdr *iph;
- int ptr;
struct sk_buff *skb2;
- unsigned int mtu, hlen, left, len, ll_rs;
- int offset;
- __be16 not_last_frag;
struct rtable *rt = skb_rtable(skb);
+ unsigned int mtu, hlen, ll_rs;
+ struct ip_fraglist_iter iter;
+ struct ip_frag_state state;
int err = 0;
/* for offloaded checksums cleanup checksum before fragmentation */
@@ -654,49 +820,24 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb,
}
/* Everything is OK. Generate! */
-
- err = 0;
- offset = 0;
- frag = skb_shinfo(skb)->frag_list;
- skb_frag_list_init(skb);
- skb->data_len = first_len - skb_headlen(skb);
- skb->len = first_len;
- iph->tot_len = htons(first_len);
- iph->frag_off = htons(IP_MF);
- ip_send_check(iph);
+ ip_fraglist_init(skb, iph, hlen, &iter);
for (;;) {
/* Prepare header of the next frame,
* before previous one went down. */
- if (frag) {
- frag->ip_summed = CHECKSUM_NONE;
- skb_reset_transport_header(frag);
- __skb_push(frag, hlen);
- skb_reset_network_header(frag);
- memcpy(skb_network_header(frag), iph, hlen);
- iph = ip_hdr(frag);
- iph->tot_len = htons(frag->len);
- ip_copy_metadata(frag, skb);
- if (offset == 0)
- ip_options_fragment(frag);
- offset += skb->len - hlen;
- iph->frag_off = htons(offset>>3);
- if (frag->next)
- iph->frag_off |= htons(IP_MF);
- /* Ready, complete checksum */
- ip_send_check(iph);
+ if (iter.frag) {
+ ip_fraglist_ipcb_prepare(skb, &iter);
+ ip_fraglist_prepare(skb, &iter);
}
err = output(net, sk, skb);
if (!err)
IP_INC_STATS(net, IPSTATS_MIB_FRAGCREATES);
- if (err || !frag)
+ if (err || !iter.frag)
break;
- skb = frag;
- frag = skb->next;
- skb_mark_not_on_list(skb);
+ skb = ip_fraglist_next(&iter);
}
if (err == 0) {
@@ -704,7 +845,7 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb,
return 0;
}
- kfree_skb_list(frag);
+ kfree_skb_list(iter.frag_list);
IP_INC_STATS(net, IPSTATS_MIB_FRAGFAILS);
return err;
@@ -720,105 +861,29 @@ slow_path_clean:
}
slow_path:
- iph = ip_hdr(skb);
-
- left = skb->len - hlen; /* Space per frame */
- ptr = hlen; /* Where to start from */
-
/*
* Fragment the datagram.
*/
- offset = (ntohs(iph->frag_off) & IP_OFFSET) << 3;
- not_last_frag = iph->frag_off & htons(IP_MF);
+ ip_frag_init(skb, hlen, ll_rs, mtu, &state);
/*
* Keep copying data until we run out.
*/
- while (left > 0) {
- len = left;
- /* IF: it doesn't fit, use 'mtu' - the data space left */
- if (len > mtu)
- len = mtu;
- /* IF: we are not sending up to and including the packet end
- then align the next start on an eight byte boundary */
- if (len < left) {
- len &= ~7;
- }
+ while (state.left > 0) {
+ bool first_frag = (state.offset == 0);
- /* Allocate buffer */
- skb2 = alloc_skb(len + hlen + ll_rs, GFP_ATOMIC);
- if (!skb2) {
- err = -ENOMEM;
+ skb2 = ip_frag_next(skb, &state);
+ if (IS_ERR(skb2)) {
+ err = PTR_ERR(skb2);
goto fail;
}
-
- /*
- * Set up data on packet
- */
-
- ip_copy_metadata(skb2, skb);
- skb_reserve(skb2, ll_rs);
- skb_put(skb2, len + hlen);
- skb_reset_network_header(skb2);
- skb2->transport_header = skb2->network_header + hlen;
-
- /*
- * Charge the memory for the fragment to any owner
- * it might possess
- */
-
- if (skb->sk)
- skb_set_owner_w(skb2, skb->sk);
-
- /*
- * Copy the packet header into the new buffer.
- */
-
- skb_copy_from_linear_data(skb, skb_network_header(skb2), hlen);
-
- /*
- * Copy a block of the IP datagram.
- */
- if (skb_copy_bits(skb, ptr, skb_transport_header(skb2), len))
- BUG();
- left -= len;
-
- /*
- * Fill in the new header fields.
- */
- iph = ip_hdr(skb2);
- iph->frag_off = htons((offset >> 3));
-
- if (IPCB(skb)->flags & IPSKB_FRAG_PMTU)
- iph->frag_off |= htons(IP_DF);
-
- /* ANK: dirty, but effective trick. Upgrade options only if
- * the segment to be fragmented was THE FIRST (otherwise,
- * options are already fixed) and make it ONCE
- * on the initial skb, so that all the following fragments
- * will inherit fixed options.
- */
- if (offset == 0)
- ip_options_fragment(skb);
-
- /*
- * Added AC : If we are fragmenting a fragment that's not the
- * last fragment then keep MF on each bit
- */
- if (left > 0 || not_last_frag)
- iph->frag_off |= htons(IP_MF);
- ptr += len;
- offset += len;
+ ip_frag_ipcb(skb, skb2, first_frag, &state);
/*
* Put this fragment into the sending queue.
*/
- iph->tot_len = htons(len + hlen);
-
- ip_send_check(iph);
-
err = output(net, sk, skb2);
if (err)
goto fail;
@@ -890,7 +955,7 @@ static int __ip_append_data(struct sock *sk,
int csummode = CHECKSUM_NONE;
struct rtable *rt = (struct rtable *)cork->dst;
unsigned int wmem_alloc_delta = 0;
- bool paged, extra_uref;
+ bool paged, extra_uref = false;
u32 tskey = 0;
skb = skb_peek_tail(queue);
@@ -930,7 +995,7 @@ static int __ip_append_data(struct sock *sk,
uarg = sock_zerocopy_realloc(sk, length, skb_zcopy(skb));
if (!uarg)
return -ENOBUFS;
- extra_uref = true;
+ extra_uref = !skb; /* only extra ref if !MSG_MORE */
if (rt->dst.dev->features & NETIF_F_SG &&
csummode == CHECKSUM_PARTIAL) {
paged = true;
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index 82f341e84fae..aa3fd61818c4 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -343,6 +343,8 @@ int ip_ra_control(struct sock *sk, unsigned char on,
return -EINVAL;
new_ra = on ? kmalloc(sizeof(*new_ra), GFP_KERNEL) : NULL;
+ if (on && !new_ra)
+ return -ENOMEM;
mutex_lock(&net->ipv4.ra_mutex);
for (rap = &net->ipv4.ra_chain;
diff --git a/net/ipv4/netfilter/nft_fib_ipv4.c b/net/ipv4/netfilter/nft_fib_ipv4.c
index 94eb25bc8d7e..c8888e52591f 100644
--- a/net/ipv4/netfilter/nft_fib_ipv4.c
+++ b/net/ipv4/netfilter/nft_fib_ipv4.c
@@ -58,11 +58,6 @@ void nft_fib4_eval_type(const struct nft_expr *expr, struct nft_regs *regs,
}
EXPORT_SYMBOL_GPL(nft_fib4_eval_type);
-static int get_ifindex(const struct net_device *dev)
-{
- return dev ? dev->ifindex : 0;
-}
-
void nft_fib4_eval(const struct nft_expr *expr, struct nft_regs *regs,
const struct nft_pktinfo *pkt)
{
@@ -94,8 +89,7 @@ void nft_fib4_eval(const struct nft_expr *expr, struct nft_regs *regs,
if (nft_hook(pkt) == NF_INET_PRE_ROUTING &&
nft_fib_is_loopback(pkt->skb, nft_in(pkt))) {
- nft_fib_store_result(dest, priv, pkt,
- nft_in(pkt)->ifindex);
+ nft_fib_store_result(dest, priv, nft_in(pkt));
return;
}
@@ -108,8 +102,7 @@ void nft_fib4_eval(const struct nft_expr *expr, struct nft_regs *regs,
if (ipv4_is_zeronet(iph->saddr)) {
if (ipv4_is_lbcast(iph->daddr) ||
ipv4_is_local_multicast(iph->daddr)) {
- nft_fib_store_result(dest, priv, pkt,
- get_ifindex(pkt->skb->dev));
+ nft_fib_store_result(dest, priv, pkt->skb->dev);
return;
}
}
@@ -150,17 +143,7 @@ void nft_fib4_eval(const struct nft_expr *expr, struct nft_regs *regs,
found = oif;
}
- switch (priv->result) {
- case NFT_FIB_RESULT_OIF:
- *dest = found->ifindex;
- break;
- case NFT_FIB_RESULT_OIFNAME:
- strncpy((char *)dest, found->name, IFNAMSIZ);
- break;
- default:
- WARN_ON_ONCE(1);
- break;
- }
+ nft_fib_store_result(dest, priv, found);
}
EXPORT_SYMBOL_GPL(nft_fib4_eval);
diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c
new file mode 100644
index 000000000000..7a5a3d08fec3
--- /dev/null
+++ b/net/ipv4/nexthop.c
@@ -0,0 +1,1479 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Generic nexthop implementation
+ *
+ * Copyright (c) 2017-19 Cumulus Networks
+ * Copyright (c) 2017-19 David Ahern <dsa@cumulusnetworks.com>
+ */
+
+#include <linux/nexthop.h>
+#include <linux/rtnetlink.h>
+#include <linux/slab.h>
+#include <net/arp.h>
+#include <net/ipv6_stubs.h>
+#include <net/lwtunnel.h>
+#include <net/ndisc.h>
+#include <net/nexthop.h>
+#include <net/route.h>
+#include <net/sock.h>
+
+static void remove_nexthop(struct net *net, struct nexthop *nh,
+ struct nl_info *nlinfo);
+
+#define NH_DEV_HASHBITS 8
+#define NH_DEV_HASHSIZE (1U << NH_DEV_HASHBITS)
+
+static const struct nla_policy rtm_nh_policy[NHA_MAX + 1] = {
+ [NHA_UNSPEC] = { .strict_start_type = NHA_UNSPEC + 1 },
+ [NHA_ID] = { .type = NLA_U32 },
+ [NHA_GROUP] = { .type = NLA_BINARY },
+ [NHA_GROUP_TYPE] = { .type = NLA_U16 },
+ [NHA_BLACKHOLE] = { .type = NLA_FLAG },
+ [NHA_OIF] = { .type = NLA_U32 },
+ [NHA_GATEWAY] = { .type = NLA_BINARY },
+ [NHA_ENCAP_TYPE] = { .type = NLA_U16 },
+ [NHA_ENCAP] = { .type = NLA_NESTED },
+ [NHA_GROUPS] = { .type = NLA_FLAG },
+ [NHA_MASTER] = { .type = NLA_U32 },
+};
+
+static unsigned int nh_dev_hashfn(unsigned int val)
+{
+ unsigned int mask = NH_DEV_HASHSIZE - 1;
+
+ return (val ^
+ (val >> NH_DEV_HASHBITS) ^
+ (val >> (NH_DEV_HASHBITS * 2))) & mask;
+}
+
+static void nexthop_devhash_add(struct net *net, struct nh_info *nhi)
+{
+ struct net_device *dev = nhi->fib_nhc.nhc_dev;
+ struct hlist_head *head;
+ unsigned int hash;
+
+ WARN_ON(!dev);
+
+ hash = nh_dev_hashfn(dev->ifindex);
+ head = &net->nexthop.devhash[hash];
+ hlist_add_head(&nhi->dev_hash, head);
+}
+
+static void nexthop_free_mpath(struct nexthop *nh)
+{
+ struct nh_group *nhg;
+ int i;
+
+ nhg = rcu_dereference_raw(nh->nh_grp);
+ for (i = 0; i < nhg->num_nh; ++i)
+ WARN_ON(nhg->nh_entries[i].nh);
+
+ kfree(nhg);
+}
+
+static void nexthop_free_single(struct nexthop *nh)
+{
+ struct nh_info *nhi;
+
+ nhi = rcu_dereference_raw(nh->nh_info);
+ switch (nhi->family) {
+ case AF_INET:
+ fib_nh_release(nh->net, &nhi->fib_nh);
+ break;
+ case AF_INET6:
+ ipv6_stub->fib6_nh_release(&nhi->fib6_nh);
+ break;
+ }
+ kfree(nhi);
+}
+
+void nexthop_free_rcu(struct rcu_head *head)
+{
+ struct nexthop *nh = container_of(head, struct nexthop, rcu);
+
+ if (nh->is_group)
+ nexthop_free_mpath(nh);
+ else
+ nexthop_free_single(nh);
+
+ kfree(nh);
+}
+EXPORT_SYMBOL_GPL(nexthop_free_rcu);
+
+static struct nexthop *nexthop_alloc(void)
+{
+ struct nexthop *nh;
+
+ nh = kzalloc(sizeof(struct nexthop), GFP_KERNEL);
+ if (nh) {
+ INIT_LIST_HEAD(&nh->grp_list);
+ }
+ return nh;
+}
+
+static struct nh_group *nexthop_grp_alloc(u16 num_nh)
+{
+ size_t sz = offsetof(struct nexthop, nh_grp)
+ + sizeof(struct nh_group)
+ + sizeof(struct nh_grp_entry) * num_nh;
+ struct nh_group *nhg;
+
+ nhg = kzalloc(sz, GFP_KERNEL);
+ if (nhg)
+ nhg->num_nh = num_nh;
+
+ return nhg;
+}
+
+static void nh_base_seq_inc(struct net *net)
+{
+ while (++net->nexthop.seq == 0)
+ ;
+}
+
+/* no reference taken; rcu lock or rtnl must be held */
+struct nexthop *nexthop_find_by_id(struct net *net, u32 id)
+{
+ struct rb_node **pp, *parent = NULL, *next;
+
+ pp = &net->nexthop.rb_root.rb_node;
+ while (1) {
+ struct nexthop *nh;
+
+ next = rcu_dereference_raw(*pp);
+ if (!next)
+ break;
+ parent = next;
+
+ nh = rb_entry(parent, struct nexthop, rb_node);
+ if (id < nh->id)
+ pp = &next->rb_left;
+ else if (id > nh->id)
+ pp = &next->rb_right;
+ else
+ return nh;
+ }
+ return NULL;
+}
+EXPORT_SYMBOL_GPL(nexthop_find_by_id);
+
+/* used for auto id allocation; called with rtnl held */
+static u32 nh_find_unused_id(struct net *net)
+{
+ u32 id_start = net->nexthop.last_id_allocated;
+
+ while (1) {
+ net->nexthop.last_id_allocated++;
+ if (net->nexthop.last_id_allocated == id_start)
+ break;
+
+ if (!nexthop_find_by_id(net, net->nexthop.last_id_allocated))
+ return net->nexthop.last_id_allocated;
+ }
+ return 0;
+}
+
+static int nla_put_nh_group(struct sk_buff *skb, struct nh_group *nhg)
+{
+ struct nexthop_grp *p;
+ size_t len = nhg->num_nh * sizeof(*p);
+ struct nlattr *nla;
+ u16 group_type = 0;
+ int i;
+
+ if (nhg->mpath)
+ group_type = NEXTHOP_GRP_TYPE_MPATH;
+
+ if (nla_put_u16(skb, NHA_GROUP_TYPE, group_type))
+ goto nla_put_failure;
+
+ nla = nla_reserve(skb, NHA_GROUP, len);
+ if (!nla)
+ goto nla_put_failure;
+
+ p = nla_data(nla);
+ for (i = 0; i < nhg->num_nh; ++i) {
+ p->id = nhg->nh_entries[i].nh->id;
+ p->weight = nhg->nh_entries[i].weight - 1;
+ p += 1;
+ }
+
+ return 0;
+
+nla_put_failure:
+ return -EMSGSIZE;
+}
+
+static int nh_fill_node(struct sk_buff *skb, struct nexthop *nh,
+ int event, u32 portid, u32 seq, unsigned int nlflags)
+{
+ struct fib6_nh *fib6_nh;
+ struct fib_nh *fib_nh;
+ struct nlmsghdr *nlh;
+ struct nh_info *nhi;
+ struct nhmsg *nhm;
+
+ nlh = nlmsg_put(skb, portid, seq, event, sizeof(*nhm), nlflags);
+ if (!nlh)
+ return -EMSGSIZE;
+
+ nhm = nlmsg_data(nlh);
+ nhm->nh_family = AF_UNSPEC;
+ nhm->nh_flags = nh->nh_flags;
+ nhm->nh_protocol = nh->protocol;
+ nhm->nh_scope = 0;
+ nhm->resvd = 0;
+
+ if (nla_put_u32(skb, NHA_ID, nh->id))
+ goto nla_put_failure;
+
+ if (nh->is_group) {
+ struct nh_group *nhg = rtnl_dereference(nh->nh_grp);
+
+ if (nla_put_nh_group(skb, nhg))
+ goto nla_put_failure;
+ goto out;
+ }
+
+ nhi = rtnl_dereference(nh->nh_info);
+ nhm->nh_family = nhi->family;
+ if (nhi->reject_nh) {
+ if (nla_put_flag(skb, NHA_BLACKHOLE))
+ goto nla_put_failure;
+ goto out;
+ } else {
+ const struct net_device *dev;
+
+ dev = nhi->fib_nhc.nhc_dev;
+ if (dev && nla_put_u32(skb, NHA_OIF, dev->ifindex))
+ goto nla_put_failure;
+ }
+
+ nhm->nh_scope = nhi->fib_nhc.nhc_scope;
+ switch (nhi->family) {
+ case AF_INET:
+ fib_nh = &nhi->fib_nh;
+ if (fib_nh->fib_nh_gw_family &&
+ nla_put_u32(skb, NHA_GATEWAY, fib_nh->fib_nh_gw4))
+ goto nla_put_failure;
+ break;
+
+ case AF_INET6:
+ fib6_nh = &nhi->fib6_nh;
+ if (fib6_nh->fib_nh_gw_family &&
+ nla_put_in6_addr(skb, NHA_GATEWAY, &fib6_nh->fib_nh_gw6))
+ goto nla_put_failure;
+ break;
+ }
+
+ if (nhi->fib_nhc.nhc_lwtstate &&
+ lwtunnel_fill_encap(skb, nhi->fib_nhc.nhc_lwtstate,
+ NHA_ENCAP, NHA_ENCAP_TYPE) < 0)
+ goto nla_put_failure;
+
+out:
+ nlmsg_end(skb, nlh);
+ return 0;
+
+nla_put_failure:
+ return -EMSGSIZE;
+}
+
+static size_t nh_nlmsg_size_grp(struct nexthop *nh)
+{
+ struct nh_group *nhg = rtnl_dereference(nh->nh_grp);
+ size_t sz = sizeof(struct nexthop_grp) * nhg->num_nh;
+
+ return nla_total_size(sz) +
+ nla_total_size(2); /* NHA_GROUP_TYPE */
+}
+
+static size_t nh_nlmsg_size_single(struct nexthop *nh)
+{
+ struct nh_info *nhi = rtnl_dereference(nh->nh_info);
+ size_t sz;
+
+ /* covers NHA_BLACKHOLE since NHA_OIF and BLACKHOLE
+ * are mutually exclusive
+ */
+ sz = nla_total_size(4); /* NHA_OIF */
+
+ switch (nhi->family) {
+ case AF_INET:
+ if (nhi->fib_nh.fib_nh_gw_family)
+ sz += nla_total_size(4); /* NHA_GATEWAY */
+ break;
+
+ case AF_INET6:
+ /* NHA_GATEWAY */
+ if (nhi->fib6_nh.fib_nh_gw_family)
+ sz += nla_total_size(sizeof(const struct in6_addr));
+ break;
+ }
+
+ if (nhi->fib_nhc.nhc_lwtstate) {
+ sz += lwtunnel_get_encap_size(nhi->fib_nhc.nhc_lwtstate);
+ sz += nla_total_size(2); /* NHA_ENCAP_TYPE */
+ }
+
+ return sz;
+}
+
+static size_t nh_nlmsg_size(struct nexthop *nh)
+{
+ size_t sz = nla_total_size(4); /* NHA_ID */
+
+ if (nh->is_group)
+ sz += nh_nlmsg_size_grp(nh);
+ else
+ sz += nh_nlmsg_size_single(nh);
+
+ return sz;
+}
+
+static void nexthop_notify(int event, struct nexthop *nh, struct nl_info *info)
+{
+ unsigned int nlflags = info->nlh ? info->nlh->nlmsg_flags : 0;
+ u32 seq = info->nlh ? info->nlh->nlmsg_seq : 0;
+ struct sk_buff *skb;
+ int err = -ENOBUFS;
+
+ skb = nlmsg_new(nh_nlmsg_size(nh), gfp_any());
+ if (!skb)
+ goto errout;
+
+ err = nh_fill_node(skb, nh, event, info->portid, seq, nlflags);
+ if (err < 0) {
+ /* -EMSGSIZE implies BUG in nh_nlmsg_size() */
+ WARN_ON(err == -EMSGSIZE);
+ kfree_skb(skb);
+ goto errout;
+ }
+
+ rtnl_notify(skb, info->nl_net, info->portid, RTNLGRP_NEXTHOP,
+ info->nlh, gfp_any());
+ return;
+errout:
+ if (err < 0)
+ rtnl_set_sk_err(info->nl_net, RTNLGRP_NEXTHOP, err);
+}
+
+static bool valid_group_nh(struct nexthop *nh, unsigned int npaths,
+ struct netlink_ext_ack *extack)
+{
+ if (nh->is_group) {
+ struct nh_group *nhg = rtnl_dereference(nh->nh_grp);
+
+ /* nested multipath (group within a group) is not
+ * supported
+ */
+ if (nhg->mpath) {
+ NL_SET_ERR_MSG(extack,
+ "Multipath group can not be a nexthop within a group");
+ return false;
+ }
+ } else {
+ struct nh_info *nhi = rtnl_dereference(nh->nh_info);
+
+ if (nhi->reject_nh && npaths > 1) {
+ NL_SET_ERR_MSG(extack,
+ "Blackhole nexthop can not be used in a group with more than 1 path");
+ return false;
+ }
+ }
+
+ return true;
+}
+
+static int nh_check_attr_group(struct net *net, struct nlattr *tb[],
+ struct netlink_ext_ack *extack)
+{
+ unsigned int len = nla_len(tb[NHA_GROUP]);
+ struct nexthop_grp *nhg;
+ unsigned int i, j;
+
+ if (len & (sizeof(struct nexthop_grp) - 1)) {
+ NL_SET_ERR_MSG(extack,
+ "Invalid length for nexthop group attribute");
+ return -EINVAL;
+ }
+
+ /* convert len to number of nexthop ids */
+ len /= sizeof(*nhg);
+
+ nhg = nla_data(tb[NHA_GROUP]);
+ for (i = 0; i < len; ++i) {
+ if (nhg[i].resvd1 || nhg[i].resvd2) {
+ NL_SET_ERR_MSG(extack, "Reserved fields in nexthop_grp must be 0");
+ return -EINVAL;
+ }
+ if (nhg[i].weight > 254) {
+ NL_SET_ERR_MSG(extack, "Invalid value for weight");
+ return -EINVAL;
+ }
+ for (j = i + 1; j < len; ++j) {
+ if (nhg[i].id == nhg[j].id) {
+ NL_SET_ERR_MSG(extack, "Nexthop id can not be used twice in a group");
+ return -EINVAL;
+ }
+ }
+ }
+
+ nhg = nla_data(tb[NHA_GROUP]);
+ for (i = 0; i < len; ++i) {
+ struct nexthop *nh;
+
+ nh = nexthop_find_by_id(net, nhg[i].id);
+ if (!nh) {
+ NL_SET_ERR_MSG(extack, "Invalid nexthop id");
+ return -EINVAL;
+ }
+ if (!valid_group_nh(nh, len, extack))
+ return -EINVAL;
+ }
+ for (i = NHA_GROUP + 1; i < __NHA_MAX; ++i) {
+ if (!tb[i])
+ continue;
+
+ NL_SET_ERR_MSG(extack,
+ "No other attributes can be set in nexthop groups");
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static bool ipv6_good_nh(const struct fib6_nh *nh)
+{
+ int state = NUD_REACHABLE;
+ struct neighbour *n;
+
+ rcu_read_lock_bh();
+
+ n = __ipv6_neigh_lookup_noref_stub(nh->fib_nh_dev, &nh->fib_nh_gw6);
+ if (n)
+ state = n->nud_state;
+
+ rcu_read_unlock_bh();
+
+ return !!(state & NUD_VALID);
+}
+
+static bool ipv4_good_nh(const struct fib_nh *nh)
+{
+ int state = NUD_REACHABLE;
+ struct neighbour *n;
+
+ rcu_read_lock_bh();
+
+ n = __ipv4_neigh_lookup_noref(nh->fib_nh_dev,
+ (__force u32)nh->fib_nh_gw4);
+ if (n)
+ state = n->nud_state;
+
+ rcu_read_unlock_bh();
+
+ return !!(state & NUD_VALID);
+}
+
+struct nexthop *nexthop_select_path(struct nexthop *nh, int hash)
+{
+ struct nexthop *rc = NULL;
+ struct nh_group *nhg;
+ int i;
+
+ if (!nh->is_group)
+ return nh;
+
+ nhg = rcu_dereference(nh->nh_grp);
+ for (i = 0; i < nhg->num_nh; ++i) {
+ struct nh_grp_entry *nhge = &nhg->nh_entries[i];
+ struct nh_info *nhi;
+
+ if (hash > atomic_read(&nhge->upper_bound))
+ continue;
+
+ /* nexthops always check if it is good and does
+ * not rely on a sysctl for this behavior
+ */
+ nhi = rcu_dereference(nhge->nh->nh_info);
+ switch (nhi->family) {
+ case AF_INET:
+ if (ipv4_good_nh(&nhi->fib_nh))
+ return nhge->nh;
+ break;
+ case AF_INET6:
+ if (ipv6_good_nh(&nhi->fib6_nh))
+ return nhge->nh;
+ break;
+ }
+
+ if (!rc)
+ rc = nhge->nh;
+ }
+
+ return rc;
+}
+EXPORT_SYMBOL_GPL(nexthop_select_path);
+
+static void nh_group_rebalance(struct nh_group *nhg)
+{
+ int total = 0;
+ int w = 0;
+ int i;
+
+ for (i = 0; i < nhg->num_nh; ++i)
+ total += nhg->nh_entries[i].weight;
+
+ for (i = 0; i < nhg->num_nh; ++i) {
+ struct nh_grp_entry *nhge = &nhg->nh_entries[i];
+ int upper_bound;
+
+ w += nhge->weight;
+ upper_bound = DIV_ROUND_CLOSEST_ULL((u64)w << 31, total) - 1;
+ atomic_set(&nhge->upper_bound, upper_bound);
+ }
+}
+
+static void remove_nh_grp_entry(struct nh_grp_entry *nhge,
+ struct nh_group *nhg,
+ struct nl_info *nlinfo)
+{
+ struct nexthop *nh = nhge->nh;
+ struct nh_grp_entry *nhges;
+ bool found = false;
+ int i;
+
+ WARN_ON(!nh);
+
+ nhges = nhg->nh_entries;
+ for (i = 0; i < nhg->num_nh; ++i) {
+ if (found) {
+ nhges[i-1].nh = nhges[i].nh;
+ nhges[i-1].weight = nhges[i].weight;
+ list_del(&nhges[i].nh_list);
+ list_add(&nhges[i-1].nh_list, &nhges[i-1].nh->grp_list);
+ } else if (nhg->nh_entries[i].nh == nh) {
+ found = true;
+ }
+ }
+
+ if (WARN_ON(!found))
+ return;
+
+ nhg->num_nh--;
+ nhg->nh_entries[nhg->num_nh].nh = NULL;
+
+ nh_group_rebalance(nhg);
+
+ nexthop_put(nh);
+
+ if (nlinfo)
+ nexthop_notify(RTM_NEWNEXTHOP, nhge->nh_parent, nlinfo);
+}
+
+static void remove_nexthop_from_groups(struct net *net, struct nexthop *nh,
+ struct nl_info *nlinfo)
+{
+ struct nh_grp_entry *nhge, *tmp;
+
+ list_for_each_entry_safe(nhge, tmp, &nh->grp_list, nh_list) {
+ struct nh_group *nhg;
+
+ list_del(&nhge->nh_list);
+ nhg = rtnl_dereference(nhge->nh_parent->nh_grp);
+ remove_nh_grp_entry(nhge, nhg, nlinfo);
+
+ /* if this group has no more entries then remove it */
+ if (!nhg->num_nh)
+ remove_nexthop(net, nhge->nh_parent, nlinfo);
+ }
+}
+
+static void remove_nexthop_group(struct nexthop *nh, struct nl_info *nlinfo)
+{
+ struct nh_group *nhg = rcu_dereference_rtnl(nh->nh_grp);
+ int i, num_nh = nhg->num_nh;
+
+ for (i = 0; i < num_nh; ++i) {
+ struct nh_grp_entry *nhge = &nhg->nh_entries[i];
+
+ if (WARN_ON(!nhge->nh))
+ continue;
+
+ list_del(&nhge->nh_list);
+ nexthop_put(nhge->nh);
+ nhge->nh = NULL;
+ nhg->num_nh--;
+ }
+}
+
+static void __remove_nexthop(struct net *net, struct nexthop *nh,
+ struct nl_info *nlinfo)
+{
+ if (nh->is_group) {
+ remove_nexthop_group(nh, nlinfo);
+ } else {
+ struct nh_info *nhi;
+
+ nhi = rtnl_dereference(nh->nh_info);
+ if (nhi->fib_nhc.nhc_dev)
+ hlist_del(&nhi->dev_hash);
+
+ remove_nexthop_from_groups(net, nh, nlinfo);
+ }
+}
+
+static void remove_nexthop(struct net *net, struct nexthop *nh,
+ struct nl_info *nlinfo)
+{
+ /* remove from the tree */
+ rb_erase(&nh->rb_node, &net->nexthop.rb_root);
+
+ if (nlinfo)
+ nexthop_notify(RTM_DELNEXTHOP, nh, nlinfo);
+
+ __remove_nexthop(net, nh, nlinfo);
+ nh_base_seq_inc(net);
+
+ nexthop_put(nh);
+}
+
+static int replace_nexthop(struct net *net, struct nexthop *old,
+ struct nexthop *new, struct netlink_ext_ack *extack)
+{
+ return -EEXIST;
+}
+
+/* called with rtnl_lock held */
+static int insert_nexthop(struct net *net, struct nexthop *new_nh,
+ struct nh_config *cfg, struct netlink_ext_ack *extack)
+{
+ struct rb_node **pp, *parent = NULL, *next;
+ struct rb_root *root = &net->nexthop.rb_root;
+ bool replace = !!(cfg->nlflags & NLM_F_REPLACE);
+ bool create = !!(cfg->nlflags & NLM_F_CREATE);
+ u32 new_id = new_nh->id;
+ int rc = -EEXIST;
+
+ pp = &root->rb_node;
+ while (1) {
+ struct nexthop *nh;
+
+ next = rtnl_dereference(*pp);
+ if (!next)
+ break;
+
+ parent = next;
+
+ nh = rb_entry(parent, struct nexthop, rb_node);
+ if (new_id < nh->id) {
+ pp = &next->rb_left;
+ } else if (new_id > nh->id) {
+ pp = &next->rb_right;
+ } else if (replace) {
+ rc = replace_nexthop(net, nh, new_nh, extack);
+ if (!rc)
+ new_nh = nh; /* send notification with old nh */
+ goto out;
+ } else {
+ /* id already exists and not a replace */
+ goto out;
+ }
+ }
+
+ if (replace && !create) {
+ NL_SET_ERR_MSG(extack, "Replace specified without create and no entry exists");
+ rc = -ENOENT;
+ goto out;
+ }
+
+ rb_link_node_rcu(&new_nh->rb_node, parent, pp);
+ rb_insert_color(&new_nh->rb_node, root);
+ rc = 0;
+out:
+ if (!rc) {
+ nh_base_seq_inc(net);
+ nexthop_notify(RTM_NEWNEXTHOP, new_nh, &cfg->nlinfo);
+ }
+
+ return rc;
+}
+
+/* rtnl */
+/* remove all nexthops tied to a device being deleted */
+static void nexthop_flush_dev(struct net_device *dev)
+{
+ unsigned int hash = nh_dev_hashfn(dev->ifindex);
+ struct net *net = dev_net(dev);
+ struct hlist_head *head = &net->nexthop.devhash[hash];
+ struct hlist_node *n;
+ struct nh_info *nhi;
+
+ hlist_for_each_entry_safe(nhi, n, head, dev_hash) {
+ if (nhi->fib_nhc.nhc_dev != dev)
+ continue;
+
+ remove_nexthop(net, nhi->nh_parent, NULL);
+ }
+}
+
+/* rtnl; called when net namespace is deleted */
+static void flush_all_nexthops(struct net *net)
+{
+ struct rb_root *root = &net->nexthop.rb_root;
+ struct rb_node *node;
+ struct nexthop *nh;
+
+ while ((node = rb_first(root))) {
+ nh = rb_entry(node, struct nexthop, rb_node);
+ remove_nexthop(net, nh, NULL);
+ cond_resched();
+ }
+}
+
+static struct nexthop *nexthop_create_group(struct net *net,
+ struct nh_config *cfg)
+{
+ struct nlattr *grps_attr = cfg->nh_grp;
+ struct nexthop_grp *entry = nla_data(grps_attr);
+ struct nh_group *nhg;
+ struct nexthop *nh;
+ int i;
+
+ nh = nexthop_alloc();
+ if (!nh)
+ return ERR_PTR(-ENOMEM);
+
+ nh->is_group = 1;
+
+ nhg = nexthop_grp_alloc(nla_len(grps_attr) / sizeof(*entry));
+ if (!nhg) {
+ kfree(nh);
+ return ERR_PTR(-ENOMEM);
+ }
+
+ for (i = 0; i < nhg->num_nh; ++i) {
+ struct nexthop *nhe;
+ struct nh_info *nhi;
+
+ nhe = nexthop_find_by_id(net, entry[i].id);
+ if (!nexthop_get(nhe))
+ goto out_no_nh;
+
+ nhi = rtnl_dereference(nhe->nh_info);
+ if (nhi->family == AF_INET)
+ nhg->has_v4 = true;
+
+ nhg->nh_entries[i].nh = nhe;
+ nhg->nh_entries[i].weight = entry[i].weight + 1;
+ list_add(&nhg->nh_entries[i].nh_list, &nhe->grp_list);
+ nhg->nh_entries[i].nh_parent = nh;
+ }
+
+ if (cfg->nh_grp_type == NEXTHOP_GRP_TYPE_MPATH) {
+ nhg->mpath = 1;
+ nh_group_rebalance(nhg);
+ }
+
+ rcu_assign_pointer(nh->nh_grp, nhg);
+
+ return nh;
+
+out_no_nh:
+ for (; i >= 0; --i)
+ nexthop_put(nhg->nh_entries[i].nh);
+
+ kfree(nhg);
+ kfree(nh);
+
+ return ERR_PTR(-ENOENT);
+}
+
+static int nh_create_ipv4(struct net *net, struct nexthop *nh,
+ struct nh_info *nhi, struct nh_config *cfg,
+ struct netlink_ext_ack *extack)
+{
+ struct fib_nh *fib_nh = &nhi->fib_nh;
+ struct fib_config fib_cfg = {
+ .fc_oif = cfg->nh_ifindex,
+ .fc_gw4 = cfg->gw.ipv4,
+ .fc_gw_family = cfg->gw.ipv4 ? AF_INET : 0,
+ .fc_flags = cfg->nh_flags,
+ .fc_encap = cfg->nh_encap,
+ .fc_encap_type = cfg->nh_encap_type,
+ };
+ u32 tb_id = l3mdev_fib_table(cfg->dev);
+ int err = -EINVAL;
+
+ err = fib_nh_init(net, fib_nh, &fib_cfg, 1, extack);
+ if (err) {
+ fib_nh_release(net, fib_nh);
+ goto out;
+ }
+
+ /* sets nh_dev if successful */
+ err = fib_check_nh(net, fib_nh, tb_id, 0, extack);
+ if (!err) {
+ nh->nh_flags = fib_nh->fib_nh_flags;
+ fib_info_update_nh_saddr(net, fib_nh, fib_nh->fib_nh_scope);
+ } else {
+ fib_nh_release(net, fib_nh);
+ }
+out:
+ return err;
+}
+
+static int nh_create_ipv6(struct net *net, struct nexthop *nh,
+ struct nh_info *nhi, struct nh_config *cfg,
+ struct netlink_ext_ack *extack)
+{
+ struct fib6_nh *fib6_nh = &nhi->fib6_nh;
+ struct fib6_config fib6_cfg = {
+ .fc_table = l3mdev_fib_table(cfg->dev),
+ .fc_ifindex = cfg->nh_ifindex,
+ .fc_gateway = cfg->gw.ipv6,
+ .fc_flags = cfg->nh_flags,
+ .fc_encap = cfg->nh_encap,
+ .fc_encap_type = cfg->nh_encap_type,
+ };
+ int err;
+
+ if (!ipv6_addr_any(&cfg->gw.ipv6))
+ fib6_cfg.fc_flags |= RTF_GATEWAY;
+
+ /* sets nh_dev if successful */
+ err = ipv6_stub->fib6_nh_init(net, fib6_nh, &fib6_cfg, GFP_KERNEL,
+ extack);
+ if (err)
+ ipv6_stub->fib6_nh_release(fib6_nh);
+ else
+ nh->nh_flags = fib6_nh->fib_nh_flags;
+
+ return err;
+}
+
+static struct nexthop *nexthop_create(struct net *net, struct nh_config *cfg,
+ struct netlink_ext_ack *extack)
+{
+ struct nh_info *nhi;
+ struct nexthop *nh;
+ int err = 0;
+
+ nh = nexthop_alloc();
+ if (!nh)
+ return ERR_PTR(-ENOMEM);
+
+ nhi = kzalloc(sizeof(*nhi), GFP_KERNEL);
+ if (!nhi) {
+ kfree(nh);
+ return ERR_PTR(-ENOMEM);
+ }
+
+ nh->nh_flags = cfg->nh_flags;
+ nh->net = net;
+
+ nhi->nh_parent = nh;
+ nhi->family = cfg->nh_family;
+ nhi->fib_nhc.nhc_scope = RT_SCOPE_LINK;
+
+ if (cfg->nh_blackhole) {
+ nhi->reject_nh = 1;
+ cfg->nh_ifindex = net->loopback_dev->ifindex;
+ }
+
+ switch (cfg->nh_family) {
+ case AF_INET:
+ err = nh_create_ipv4(net, nh, nhi, cfg, extack);
+ break;
+ case AF_INET6:
+ err = nh_create_ipv6(net, nh, nhi, cfg, extack);
+ break;
+ }
+
+ if (err) {
+ kfree(nhi);
+ kfree(nh);
+ return ERR_PTR(err);
+ }
+
+ /* add the entry to the device based hash */
+ nexthop_devhash_add(net, nhi);
+
+ rcu_assign_pointer(nh->nh_info, nhi);
+
+ return nh;
+}
+
+/* called with rtnl lock held */
+static struct nexthop *nexthop_add(struct net *net, struct nh_config *cfg,
+ struct netlink_ext_ack *extack)
+{
+ struct nexthop *nh;
+ int err;
+
+ if (cfg->nlflags & NLM_F_REPLACE && !cfg->nh_id) {
+ NL_SET_ERR_MSG(extack, "Replace requires nexthop id");
+ return ERR_PTR(-EINVAL);
+ }
+
+ if (!cfg->nh_id) {
+ cfg->nh_id = nh_find_unused_id(net);
+ if (!cfg->nh_id) {
+ NL_SET_ERR_MSG(extack, "No unused id");
+ return ERR_PTR(-EINVAL);
+ }
+ }
+
+ if (cfg->nh_grp)
+ nh = nexthop_create_group(net, cfg);
+ else
+ nh = nexthop_create(net, cfg, extack);
+
+ if (IS_ERR(nh))
+ return nh;
+
+ refcount_set(&nh->refcnt, 1);
+ nh->id = cfg->nh_id;
+ nh->protocol = cfg->nh_protocol;
+ nh->net = net;
+
+ err = insert_nexthop(net, nh, cfg, extack);
+ if (err) {
+ __remove_nexthop(net, nh, NULL);
+ nexthop_put(nh);
+ nh = ERR_PTR(err);
+ }
+
+ return nh;
+}
+
+static int rtm_to_nh_config(struct net *net, struct sk_buff *skb,
+ struct nlmsghdr *nlh, struct nh_config *cfg,
+ struct netlink_ext_ack *extack)
+{
+ struct nhmsg *nhm = nlmsg_data(nlh);
+ struct nlattr *tb[NHA_MAX + 1];
+ int err;
+
+ err = nlmsg_parse(nlh, sizeof(*nhm), tb, NHA_MAX, rtm_nh_policy,
+ extack);
+ if (err < 0)
+ return err;
+
+ err = -EINVAL;
+ if (nhm->resvd || nhm->nh_scope) {
+ NL_SET_ERR_MSG(extack, "Invalid values in ancillary header");
+ goto out;
+ }
+ if (nhm->nh_flags & ~NEXTHOP_VALID_USER_FLAGS) {
+ NL_SET_ERR_MSG(extack, "Invalid nexthop flags in ancillary header");
+ goto out;
+ }
+
+ switch (nhm->nh_family) {
+ case AF_INET:
+ case AF_INET6:
+ break;
+ case AF_UNSPEC:
+ if (tb[NHA_GROUP])
+ break;
+ /* fallthrough */
+ default:
+ NL_SET_ERR_MSG(extack, "Invalid address family");
+ goto out;
+ }
+
+ if (tb[NHA_GROUPS] || tb[NHA_MASTER]) {
+ NL_SET_ERR_MSG(extack, "Invalid attributes in request");
+ goto out;
+ }
+
+ memset(cfg, 0, sizeof(*cfg));
+ cfg->nlflags = nlh->nlmsg_flags;
+ cfg->nlinfo.portid = NETLINK_CB(skb).portid;
+ cfg->nlinfo.nlh = nlh;
+ cfg->nlinfo.nl_net = net;
+
+ cfg->nh_family = nhm->nh_family;
+ cfg->nh_protocol = nhm->nh_protocol;
+ cfg->nh_flags = nhm->nh_flags;
+
+ if (tb[NHA_ID])
+ cfg->nh_id = nla_get_u32(tb[NHA_ID]);
+
+ if (tb[NHA_GROUP]) {
+ if (nhm->nh_family != AF_UNSPEC) {
+ NL_SET_ERR_MSG(extack, "Invalid family for group");
+ goto out;
+ }
+ cfg->nh_grp = tb[NHA_GROUP];
+
+ cfg->nh_grp_type = NEXTHOP_GRP_TYPE_MPATH;
+ if (tb[NHA_GROUP_TYPE])
+ cfg->nh_grp_type = nla_get_u16(tb[NHA_GROUP_TYPE]);
+
+ if (cfg->nh_grp_type > NEXTHOP_GRP_TYPE_MAX) {
+ NL_SET_ERR_MSG(extack, "Invalid group type");
+ goto out;
+ }
+ err = nh_check_attr_group(net, tb, extack);
+
+ /* no other attributes should be set */
+ goto out;
+ }
+
+ if (tb[NHA_BLACKHOLE]) {
+ if (tb[NHA_GATEWAY] || tb[NHA_OIF] ||
+ tb[NHA_ENCAP] || tb[NHA_ENCAP_TYPE]) {
+ NL_SET_ERR_MSG(extack, "Blackhole attribute can not be used with gateway or oif");
+ goto out;
+ }
+
+ cfg->nh_blackhole = 1;
+ err = 0;
+ goto out;
+ }
+
+ if (!tb[NHA_OIF]) {
+ NL_SET_ERR_MSG(extack, "Device attribute required for non-blackhole nexthops");
+ goto out;
+ }
+
+ cfg->nh_ifindex = nla_get_u32(tb[NHA_OIF]);
+ if (cfg->nh_ifindex)
+ cfg->dev = __dev_get_by_index(net, cfg->nh_ifindex);
+
+ if (!cfg->dev) {
+ NL_SET_ERR_MSG(extack, "Invalid device index");
+ goto out;
+ } else if (!(cfg->dev->flags & IFF_UP)) {
+ NL_SET_ERR_MSG(extack, "Nexthop device is not up");
+ err = -ENETDOWN;
+ goto out;
+ } else if (!netif_carrier_ok(cfg->dev)) {
+ NL_SET_ERR_MSG(extack, "Carrier for nexthop device is down");
+ err = -ENETDOWN;
+ goto out;
+ }
+
+ err = -EINVAL;
+ if (tb[NHA_GATEWAY]) {
+ struct nlattr *gwa = tb[NHA_GATEWAY];
+
+ switch (cfg->nh_family) {
+ case AF_INET:
+ if (nla_len(gwa) != sizeof(u32)) {
+ NL_SET_ERR_MSG(extack, "Invalid gateway");
+ goto out;
+ }
+ cfg->gw.ipv4 = nla_get_be32(gwa);
+ break;
+ case AF_INET6:
+ if (nla_len(gwa) != sizeof(struct in6_addr)) {
+ NL_SET_ERR_MSG(extack, "Invalid gateway");
+ goto out;
+ }
+ cfg->gw.ipv6 = nla_get_in6_addr(gwa);
+ break;
+ default:
+ NL_SET_ERR_MSG(extack,
+ "Unknown address family for gateway");
+ goto out;
+ }
+ } else {
+ /* device only nexthop (no gateway) */
+ if (cfg->nh_flags & RTNH_F_ONLINK) {
+ NL_SET_ERR_MSG(extack,
+ "ONLINK flag can not be set for nexthop without a gateway");
+ goto out;
+ }
+ }
+
+ if (tb[NHA_ENCAP]) {
+ cfg->nh_encap = tb[NHA_ENCAP];
+
+ if (!tb[NHA_ENCAP_TYPE]) {
+ NL_SET_ERR_MSG(extack, "LWT encapsulation type is missing");
+ goto out;
+ }
+
+ cfg->nh_encap_type = nla_get_u16(tb[NHA_ENCAP_TYPE]);
+ err = lwtunnel_valid_encap_type(cfg->nh_encap_type, extack);
+ if (err < 0)
+ goto out;
+
+ } else if (tb[NHA_ENCAP_TYPE]) {
+ NL_SET_ERR_MSG(extack, "LWT encapsulation attribute is missing");
+ goto out;
+ }
+
+
+ err = 0;
+out:
+ return err;
+}
+
+/* rtnl */
+static int rtm_new_nexthop(struct sk_buff *skb, struct nlmsghdr *nlh,
+ struct netlink_ext_ack *extack)
+{
+ struct net *net = sock_net(skb->sk);
+ struct nh_config cfg;
+ struct nexthop *nh;
+ int err;
+
+ err = rtm_to_nh_config(net, skb, nlh, &cfg, extack);
+ if (!err) {
+ nh = nexthop_add(net, &cfg, extack);
+ if (IS_ERR(nh))
+ err = PTR_ERR(nh);
+ }
+
+ return err;
+}
+
+static int nh_valid_get_del_req(struct nlmsghdr *nlh, u32 *id,
+ struct netlink_ext_ack *extack)
+{
+ struct nhmsg *nhm = nlmsg_data(nlh);
+ struct nlattr *tb[NHA_MAX + 1];
+ int err, i;
+
+ err = nlmsg_parse(nlh, sizeof(*nhm), tb, NHA_MAX, rtm_nh_policy,
+ extack);
+ if (err < 0)
+ return err;
+
+ err = -EINVAL;
+ for (i = 0; i < __NHA_MAX; ++i) {
+ if (!tb[i])
+ continue;
+
+ switch (i) {
+ case NHA_ID:
+ break;
+ default:
+ NL_SET_ERR_MSG_ATTR(extack, tb[i],
+ "Unexpected attribute in request");
+ goto out;
+ }
+ }
+ if (nhm->nh_protocol || nhm->resvd || nhm->nh_scope || nhm->nh_flags) {
+ NL_SET_ERR_MSG(extack, "Invalid values in header");
+ goto out;
+ }
+
+ if (!tb[NHA_ID]) {
+ NL_SET_ERR_MSG(extack, "Nexthop id is missing");
+ goto out;
+ }
+
+ *id = nla_get_u32(tb[NHA_ID]);
+ if (!(*id))
+ NL_SET_ERR_MSG(extack, "Invalid nexthop id");
+ else
+ err = 0;
+out:
+ return err;
+}
+
+/* rtnl */
+static int rtm_del_nexthop(struct sk_buff *skb, struct nlmsghdr *nlh,
+ struct netlink_ext_ack *extack)
+{
+ struct net *net = sock_net(skb->sk);
+ struct nl_info nlinfo = {
+ .nlh = nlh,
+ .nl_net = net,
+ .portid = NETLINK_CB(skb).portid,
+ };
+ struct nexthop *nh;
+ int err;
+ u32 id;
+
+ err = nh_valid_get_del_req(nlh, &id, extack);
+ if (err)
+ return err;
+
+ nh = nexthop_find_by_id(net, id);
+ if (!nh)
+ return -ENOENT;
+
+ remove_nexthop(net, nh, &nlinfo);
+
+ return 0;
+}
+
+/* rtnl */
+static int rtm_get_nexthop(struct sk_buff *in_skb, struct nlmsghdr *nlh,
+ struct netlink_ext_ack *extack)
+{
+ struct net *net = sock_net(in_skb->sk);
+ struct sk_buff *skb = NULL;
+ struct nexthop *nh;
+ int err;
+ u32 id;
+
+ err = nh_valid_get_del_req(nlh, &id, extack);
+ if (err)
+ return err;
+
+ err = -ENOBUFS;
+ skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
+ if (!skb)
+ goto out;
+
+ err = -ENOENT;
+ nh = nexthop_find_by_id(net, id);
+ if (!nh)
+ goto errout_free;
+
+ err = nh_fill_node(skb, nh, RTM_NEWNEXTHOP, NETLINK_CB(in_skb).portid,
+ nlh->nlmsg_seq, 0);
+ if (err < 0) {
+ WARN_ON(err == -EMSGSIZE);
+ goto errout_free;
+ }
+
+ err = rtnl_unicast(skb, net, NETLINK_CB(in_skb).portid);
+out:
+ return err;
+errout_free:
+ kfree_skb(skb);
+ goto out;
+}
+
+static bool nh_dump_filtered(struct nexthop *nh, int dev_idx, int master_idx,
+ bool group_filter, u8 family)
+{
+ const struct net_device *dev;
+ const struct nh_info *nhi;
+
+ if (group_filter && !nh->is_group)
+ return true;
+
+ if (!dev_idx && !master_idx && !family)
+ return false;
+
+ if (nh->is_group)
+ return true;
+
+ nhi = rtnl_dereference(nh->nh_info);
+ if (family && nhi->family != family)
+ return true;
+
+ dev = nhi->fib_nhc.nhc_dev;
+ if (dev_idx && (!dev || dev->ifindex != dev_idx))
+ return true;
+
+ if (master_idx) {
+ struct net_device *master;
+
+ if (!dev)
+ return true;
+
+ master = netdev_master_upper_dev_get((struct net_device *)dev);
+ if (!master || master->ifindex != master_idx)
+ return true;
+ }
+
+ return false;
+}
+
+static int nh_valid_dump_req(const struct nlmsghdr *nlh, int *dev_idx,
+ int *master_idx, bool *group_filter,
+ struct netlink_callback *cb)
+{
+ struct netlink_ext_ack *extack = cb->extack;
+ struct nlattr *tb[NHA_MAX + 1];
+ struct nhmsg *nhm;
+ int err, i;
+ u32 idx;
+
+ err = nlmsg_parse(nlh, sizeof(*nhm), tb, NHA_MAX, rtm_nh_policy,
+ NULL);
+ if (err < 0)
+ return err;
+
+ for (i = 0; i <= NHA_MAX; ++i) {
+ if (!tb[i])
+ continue;
+
+ switch (i) {
+ case NHA_OIF:
+ idx = nla_get_u32(tb[i]);
+ if (idx > INT_MAX) {
+ NL_SET_ERR_MSG(extack, "Invalid device index");
+ return -EINVAL;
+ }
+ *dev_idx = idx;
+ break;
+ case NHA_MASTER:
+ idx = nla_get_u32(tb[i]);
+ if (idx > INT_MAX) {
+ NL_SET_ERR_MSG(extack, "Invalid master device index");
+ return -EINVAL;
+ }
+ *master_idx = idx;
+ break;
+ case NHA_GROUPS:
+ *group_filter = true;
+ break;
+ default:
+ NL_SET_ERR_MSG(extack, "Unsupported attribute in dump request");
+ return -EINVAL;
+ }
+ }
+
+ nhm = nlmsg_data(nlh);
+ if (nhm->nh_protocol || nhm->resvd || nhm->nh_scope || nhm->nh_flags) {
+ NL_SET_ERR_MSG(extack, "Invalid values in header for nexthop dump request");
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+/* rtnl */
+static int rtm_dump_nexthop(struct sk_buff *skb, struct netlink_callback *cb)
+{
+ struct nhmsg *nhm = nlmsg_data(cb->nlh);
+ int dev_filter_idx = 0, master_idx = 0;
+ struct net *net = sock_net(skb->sk);
+ struct rb_root *root = &net->nexthop.rb_root;
+ bool group_filter = false;
+ struct rb_node *node;
+ int idx = 0, s_idx;
+ int err;
+
+ err = nh_valid_dump_req(cb->nlh, &dev_filter_idx, &master_idx,
+ &group_filter, cb);
+ if (err < 0)
+ return err;
+
+ s_idx = cb->args[0];
+ for (node = rb_first(root); node; node = rb_next(node)) {
+ struct nexthop *nh;
+
+ if (idx < s_idx)
+ goto cont;
+
+ nh = rb_entry(node, struct nexthop, rb_node);
+ if (nh_dump_filtered(nh, dev_filter_idx, master_idx,
+ group_filter, nhm->nh_family))
+ goto cont;
+
+ err = nh_fill_node(skb, nh, RTM_NEWNEXTHOP,
+ NETLINK_CB(cb->skb).portid,
+ cb->nlh->nlmsg_seq, NLM_F_MULTI);
+ if (err < 0) {
+ if (likely(skb->len))
+ goto out;
+
+ goto out_err;
+ }
+cont:
+ idx++;
+ }
+
+out:
+ err = skb->len;
+out_err:
+ cb->args[0] = idx;
+ cb->seq = net->nexthop.seq;
+ nl_dump_check_consistent(cb, nlmsg_hdr(skb));
+
+ return err;
+}
+
+static void nexthop_sync_mtu(struct net_device *dev, u32 orig_mtu)
+{
+ unsigned int hash = nh_dev_hashfn(dev->ifindex);
+ struct net *net = dev_net(dev);
+ struct hlist_head *head = &net->nexthop.devhash[hash];
+ struct hlist_node *n;
+ struct nh_info *nhi;
+
+ hlist_for_each_entry_safe(nhi, n, head, dev_hash) {
+ if (nhi->fib_nhc.nhc_dev == dev) {
+ if (nhi->family == AF_INET)
+ fib_nhc_update_mtu(&nhi->fib_nhc, dev->mtu,
+ orig_mtu);
+ }
+ }
+}
+
+/* rtnl */
+static int nh_netdev_event(struct notifier_block *this,
+ unsigned long event, void *ptr)
+{
+ struct net_device *dev = netdev_notifier_info_to_dev(ptr);
+ struct netdev_notifier_info_ext *info_ext;
+
+ switch (event) {
+ case NETDEV_DOWN:
+ case NETDEV_UNREGISTER:
+ nexthop_flush_dev(dev);
+ break;
+ case NETDEV_CHANGE:
+ if (!(dev_get_flags(dev) & (IFF_RUNNING | IFF_LOWER_UP)))
+ nexthop_flush_dev(dev);
+ break;
+ case NETDEV_CHANGEMTU:
+ info_ext = ptr;
+ nexthop_sync_mtu(dev, info_ext->ext.mtu);
+ rt_cache_flush(dev_net(dev));
+ break;
+ }
+ return NOTIFY_DONE;
+}
+
+static struct notifier_block nh_netdev_notifier = {
+ .notifier_call = nh_netdev_event,
+};
+
+static void __net_exit nexthop_net_exit(struct net *net)
+{
+ rtnl_lock();
+ flush_all_nexthops(net);
+ rtnl_unlock();
+ kfree(net->nexthop.devhash);
+}
+
+static int __net_init nexthop_net_init(struct net *net)
+{
+ size_t sz = sizeof(struct hlist_head) * NH_DEV_HASHSIZE;
+
+ net->nexthop.rb_root = RB_ROOT;
+ net->nexthop.devhash = kzalloc(sz, GFP_KERNEL);
+ if (!net->nexthop.devhash)
+ return -ENOMEM;
+
+ return 0;
+}
+
+static struct pernet_operations nexthop_net_ops = {
+ .init = nexthop_net_init,
+ .exit = nexthop_net_exit,
+};
+
+static int __init nexthop_init(void)
+{
+ register_pernet_subsys(&nexthop_net_ops);
+
+ register_netdevice_notifier(&nh_netdev_notifier);
+
+ rtnl_register(PF_UNSPEC, RTM_NEWNEXTHOP, rtm_new_nexthop, NULL, 0);
+ rtnl_register(PF_UNSPEC, RTM_DELNEXTHOP, rtm_del_nexthop, NULL, 0);
+ rtnl_register(PF_UNSPEC, RTM_GETNEXTHOP, rtm_get_nexthop,
+ rtm_dump_nexthop, 0);
+
+ rtnl_register(PF_INET, RTM_NEWNEXTHOP, rtm_new_nexthop, NULL, 0);
+ rtnl_register(PF_INET, RTM_GETNEXTHOP, NULL, rtm_dump_nexthop, 0);
+
+ rtnl_register(PF_INET6, RTM_NEWNEXTHOP, rtm_new_nexthop, NULL, 0);
+ rtnl_register(PF_INET6, RTM_GETNEXTHOP, NULL, rtm_dump_nexthop, 0);
+
+ return 0;
+}
+subsys_initcall(nexthop_init);
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index c3610b37bb4c..4746f963c439 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -72,8 +72,8 @@ static int sockstat_seq_show(struct seq_file *seq, void *v)
seq_printf(seq, "RAW: inuse %d\n",
sock_prot_inuse_get(net, &raw_prot));
seq_printf(seq, "FRAG: inuse %u memory %lu\n",
- atomic_read(&net->ipv4.frags.rhashtable.nelems),
- frag_mem_limit(&net->ipv4.frags));
+ atomic_read(&net->ipv4.fqdir->rhashtable.nelems),
+ frag_mem_limit(net->ipv4.fqdir));
return 0;
}
@@ -291,6 +291,7 @@ static const struct snmp_mib snmp4_net_list[] = {
SNMP_MIB_ITEM("TCPAckCompressed", LINUX_MIB_TCPACKCOMPRESSED),
SNMP_MIB_ITEM("TCPZeroWindowDrop", LINUX_MIB_TCPZEROWINDOWDROP),
SNMP_MIB_ITEM("TCPRcvQDrop", LINUX_MIB_TCPRCVQDROP),
+ SNMP_MIB_ITEM("TCPFastOpenPassiveAltKey", LINUX_MIB_TCPFASTOPENPASSIVEALTKEY),
SNMP_MIB_SENTINEL
};
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 875867b64d6a..90f09e47198b 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -277,55 +277,97 @@ static int proc_allowed_congestion_control(struct ctl_table *ctl,
return ret;
}
+static int sscanf_key(char *buf, __le32 *key)
+{
+ u32 user_key[4];
+ int i, ret = 0;
+
+ if (sscanf(buf, "%x-%x-%x-%x", user_key, user_key + 1,
+ user_key + 2, user_key + 3) != 4) {
+ ret = -EINVAL;
+ } else {
+ for (i = 0; i < ARRAY_SIZE(user_key); i++)
+ key[i] = cpu_to_le32(user_key[i]);
+ }
+ pr_debug("proc TFO key set 0x%x-%x-%x-%x <- 0x%s: %u\n",
+ user_key[0], user_key[1], user_key[2], user_key[3], buf, ret);
+
+ return ret;
+}
+
static int proc_tcp_fastopen_key(struct ctl_table *table, int write,
void __user *buffer, size_t *lenp,
loff_t *ppos)
{
struct net *net = container_of(table->data, struct net,
ipv4.sysctl_tcp_fastopen);
- struct ctl_table tbl = { .maxlen = (TCP_FASTOPEN_KEY_LENGTH * 2 + 10) };
- struct tcp_fastopen_context *ctxt;
- u32 user_key[4]; /* 16 bytes, matching TCP_FASTOPEN_KEY_LENGTH */
- __le32 key[4];
- int ret, i;
+ /* maxlen to print the list of keys in hex (*2), with dashes
+ * separating doublewords and a comma in between keys.
+ */
+ struct ctl_table tbl = { .maxlen = ((TCP_FASTOPEN_KEY_LENGTH *
+ 2 * TCP_FASTOPEN_KEY_MAX) +
+ (TCP_FASTOPEN_KEY_MAX * 5)) };
+ struct tcp_fastopen_context *ctx;
+ u32 user_key[TCP_FASTOPEN_KEY_MAX * 4];
+ __le32 key[TCP_FASTOPEN_KEY_MAX * 4];
+ char *backup_data;
+ int ret, i = 0, off = 0, n_keys = 0;
tbl.data = kmalloc(tbl.maxlen, GFP_KERNEL);
if (!tbl.data)
return -ENOMEM;
rcu_read_lock();
- ctxt = rcu_dereference(net->ipv4.tcp_fastopen_ctx);
- if (ctxt)
- memcpy(key, ctxt->key, TCP_FASTOPEN_KEY_LENGTH);
- else
- memset(key, 0, sizeof(key));
+ ctx = rcu_dereference(net->ipv4.tcp_fastopen_ctx);
+ if (ctx) {
+ n_keys = tcp_fastopen_context_len(ctx);
+ memcpy(&key[0], &ctx->key[0], TCP_FASTOPEN_KEY_LENGTH * n_keys);
+ }
rcu_read_unlock();
- for (i = 0; i < ARRAY_SIZE(key); i++)
+ if (!n_keys) {
+ memset(&key[0], 0, TCP_FASTOPEN_KEY_LENGTH);
+ n_keys = 1;
+ }
+
+ for (i = 0; i < n_keys * 4; i++)
user_key[i] = le32_to_cpu(key[i]);
- snprintf(tbl.data, tbl.maxlen, "%08x-%08x-%08x-%08x",
- user_key[0], user_key[1], user_key[2], user_key[3]);
+ for (i = 0; i < n_keys; i++) {
+ off += snprintf(tbl.data + off, tbl.maxlen - off,
+ "%08x-%08x-%08x-%08x",
+ user_key[i * 4],
+ user_key[i * 4 + 1],
+ user_key[i * 4 + 2],
+ user_key[i * 4 + 3]);
+ if (i + 1 < n_keys)
+ off += snprintf(tbl.data + off, tbl.maxlen - off, ",");
+ }
+
ret = proc_dostring(&tbl, write, buffer, lenp, ppos);
if (write && ret == 0) {
- if (sscanf(tbl.data, "%x-%x-%x-%x", user_key, user_key + 1,
- user_key + 2, user_key + 3) != 4) {
+ backup_data = strchr(tbl.data, ',');
+ if (backup_data) {
+ *backup_data = '\0';
+ backup_data++;
+ }
+ if (sscanf_key(tbl.data, key)) {
ret = -EINVAL;
goto bad_key;
}
-
- for (i = 0; i < ARRAY_SIZE(user_key); i++)
- key[i] = cpu_to_le32(user_key[i]);
-
+ if (backup_data) {
+ if (sscanf_key(backup_data, key + 4)) {
+ ret = -EINVAL;
+ goto bad_key;
+ }
+ }
tcp_fastopen_reset_cipher(net, NULL, key,
+ backup_data ? key + 4 : NULL,
TCP_FASTOPEN_KEY_LENGTH);
}
bad_key:
- pr_debug("proc FO key set 0x%x-%x-%x-%x <- 0x%s: %u\n",
- user_key[0], user_key[1], user_key[2], user_key[3],
- (char *)tbl.data, ret);
kfree(tbl.data);
return ret;
}
@@ -933,7 +975,12 @@ static struct ctl_table ipv4_net_table[] = {
.procname = "tcp_fastopen_key",
.mode = 0600,
.data = &init_net.ipv4.sysctl_tcp_fastopen,
- .maxlen = ((TCP_FASTOPEN_KEY_LENGTH * 2) + 10),
+ /* maxlen to print the list of keys in hex (*2), with dashes
+ * separating doublewords and a comma in between keys.
+ */
+ .maxlen = ((TCP_FASTOPEN_KEY_LENGTH *
+ 2 * TCP_FASTOPEN_KEY_MAX) +
+ (TCP_FASTOPEN_KEY_MAX * 5)),
.proc_handler = proc_tcp_fastopen_key,
},
{
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 53d61ca3ac4b..27ce13ece510 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2790,15 +2790,24 @@ static int do_tcp_setsockopt(struct sock *sk, int level,
return err;
}
case TCP_FASTOPEN_KEY: {
- __u8 key[TCP_FASTOPEN_KEY_LENGTH];
+ __u8 key[TCP_FASTOPEN_KEY_BUF_LENGTH];
+ __u8 *backup_key = NULL;
- if (optlen != sizeof(key))
+ /* Allow a backup key as well to facilitate key rotation
+ * First key is the active one.
+ */
+ if (optlen != TCP_FASTOPEN_KEY_LENGTH &&
+ optlen != TCP_FASTOPEN_KEY_BUF_LENGTH)
return -EINVAL;
if (copy_from_user(key, optval, optlen))
return -EFAULT;
- return tcp_fastopen_reset_cipher(net, sk, key, sizeof(key));
+ if (optlen == TCP_FASTOPEN_KEY_BUF_LENGTH)
+ backup_key = key + TCP_FASTOPEN_KEY_LENGTH;
+
+ return tcp_fastopen_reset_cipher(net, sk, key, backup_key,
+ TCP_FASTOPEN_KEY_LENGTH);
}
default:
/* fallthru */
@@ -3452,21 +3461,23 @@ static int do_tcp_getsockopt(struct sock *sk, int level,
return 0;
case TCP_FASTOPEN_KEY: {
- __u8 key[TCP_FASTOPEN_KEY_LENGTH];
+ __u8 key[TCP_FASTOPEN_KEY_BUF_LENGTH];
struct tcp_fastopen_context *ctx;
+ unsigned int key_len = 0;
if (get_user(len, optlen))
return -EFAULT;
rcu_read_lock();
ctx = rcu_dereference(icsk->icsk_accept_queue.fastopenq.ctx);
- if (ctx)
- memcpy(key, ctx->key, sizeof(key));
- else
- len = 0;
+ if (ctx) {
+ key_len = tcp_fastopen_context_len(ctx) *
+ TCP_FASTOPEN_KEY_LENGTH;
+ memcpy(&key[0], &ctx->key[0], key_len);
+ }
rcu_read_unlock();
- len = min_t(unsigned int, len, sizeof(key));
+ len = min_t(unsigned int, len, key_len);
if (put_user(len, optlen))
return -EFAULT;
if (copy_to_user(optval, key, len))
diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c
index 018a48477355..8e1580485c9e 100644
--- a/net/ipv4/tcp_fastopen.c
+++ b/net/ipv4/tcp_fastopen.c
@@ -30,14 +30,20 @@ void tcp_fastopen_init_key_once(struct net *net)
* for a valid cookie, so this is an acceptable risk.
*/
get_random_bytes(key, sizeof(key));
- tcp_fastopen_reset_cipher(net, NULL, key, sizeof(key));
+ tcp_fastopen_reset_cipher(net, NULL, key, NULL, sizeof(key));
}
static void tcp_fastopen_ctx_free(struct rcu_head *head)
{
struct tcp_fastopen_context *ctx =
container_of(head, struct tcp_fastopen_context, rcu);
- crypto_free_cipher(ctx->tfm);
+ int i;
+
+ /* We own ctx, thus no need to hold the Fastopen-lock */
+ for (i = 0; i < TCP_FASTOPEN_KEY_MAX; i++) {
+ if (ctx->tfm[i])
+ crypto_free_cipher(ctx->tfm[i]);
+ }
kfree(ctx);
}
@@ -66,33 +72,54 @@ void tcp_fastopen_ctx_destroy(struct net *net)
call_rcu(&ctxt->rcu, tcp_fastopen_ctx_free);
}
+struct tcp_fastopen_context *tcp_fastopen_alloc_ctx(void *primary_key,
+ void *backup_key,
+ unsigned int len)
+{
+ struct tcp_fastopen_context *new_ctx;
+ void *key = primary_key;
+ int err, i;
+
+ new_ctx = kmalloc(sizeof(*new_ctx), GFP_KERNEL);
+ if (!new_ctx)
+ return ERR_PTR(-ENOMEM);
+ for (i = 0; i < TCP_FASTOPEN_KEY_MAX; i++)
+ new_ctx->tfm[i] = NULL;
+ for (i = 0; i < (backup_key ? 2 : 1); i++) {
+ new_ctx->tfm[i] = crypto_alloc_cipher("aes", 0, 0);
+ if (IS_ERR(new_ctx->tfm[i])) {
+ err = PTR_ERR(new_ctx->tfm[i]);
+ new_ctx->tfm[i] = NULL;
+ pr_err("TCP: TFO aes cipher alloc error: %d\n", err);
+ goto out;
+ }
+ err = crypto_cipher_setkey(new_ctx->tfm[i], key, len);
+ if (err) {
+ pr_err("TCP: TFO cipher key error: %d\n", err);
+ goto out;
+ }
+ memcpy(&new_ctx->key[i * TCP_FASTOPEN_KEY_LENGTH], key, len);
+ key = backup_key;
+ }
+ return new_ctx;
+out:
+ tcp_fastopen_ctx_free(&new_ctx->rcu);
+ return ERR_PTR(err);
+}
+
int tcp_fastopen_reset_cipher(struct net *net, struct sock *sk,
- void *key, unsigned int len)
+ void *primary_key, void *backup_key,
+ unsigned int len)
{
struct tcp_fastopen_context *ctx, *octx;
struct fastopen_queue *q;
- int err;
+ int err = 0;
- ctx = kmalloc(sizeof(*ctx), GFP_KERNEL);
- if (!ctx)
- return -ENOMEM;
- ctx->tfm = crypto_alloc_cipher("aes", 0, 0);
-
- if (IS_ERR(ctx->tfm)) {
- err = PTR_ERR(ctx->tfm);
-error: kfree(ctx);
- pr_err("TCP: TFO aes cipher alloc error: %d\n", err);
- return err;
+ ctx = tcp_fastopen_alloc_ctx(primary_key, backup_key, len);
+ if (IS_ERR(ctx)) {
+ err = PTR_ERR(ctx);
+ goto out;
}
- err = crypto_cipher_setkey(ctx->tfm, key, len);
- if (err) {
- pr_err("TCP: TFO cipher key error: %d\n", err);
- crypto_free_cipher(ctx->tfm);
- goto error;
- }
- memcpy(ctx->key, key, len);
-
-
spin_lock(&net->ipv4.tcp_fastopen_ctx_lock);
if (sk) {
q = &inet_csk(sk)->icsk_accept_queue.fastopenq;
@@ -108,28 +135,42 @@ error: kfree(ctx);
if (octx)
call_rcu(&octx->rcu, tcp_fastopen_ctx_free);
+out:
return err;
}
-static bool __tcp_fastopen_cookie_gen(struct sock *sk, const void *path,
- struct tcp_fastopen_cookie *foc)
+static bool __tcp_fastopen_cookie_gen_cipher(struct request_sock *req,
+ struct sk_buff *syn,
+ struct crypto_cipher *tfm,
+ struct tcp_fastopen_cookie *foc)
{
- struct tcp_fastopen_context *ctx;
- bool ok = false;
-
- rcu_read_lock();
+ if (req->rsk_ops->family == AF_INET) {
+ const struct iphdr *iph = ip_hdr(syn);
+ __be32 path[4] = { iph->saddr, iph->daddr, 0, 0 };
- ctx = rcu_dereference(inet_csk(sk)->icsk_accept_queue.fastopenq.ctx);
- if (!ctx)
- ctx = rcu_dereference(sock_net(sk)->ipv4.tcp_fastopen_ctx);
+ crypto_cipher_encrypt_one(tfm, foc->val, (void *)path);
+ foc->len = TCP_FASTOPEN_COOKIE_SIZE;
+ return true;
+ }
- if (ctx) {
- crypto_cipher_encrypt_one(ctx->tfm, foc->val, path);
+#if IS_ENABLED(CONFIG_IPV6)
+ if (req->rsk_ops->family == AF_INET6) {
+ const struct ipv6hdr *ip6h = ipv6_hdr(syn);
+ struct tcp_fastopen_cookie tmp;
+ struct in6_addr *buf;
+ int i;
+
+ crypto_cipher_encrypt_one(tfm, tmp.val,
+ (void *)&ip6h->saddr);
+ buf = &tmp.addr;
+ for (i = 0; i < 4; i++)
+ buf->s6_addr32[i] ^= ip6h->daddr.s6_addr32[i];
+ crypto_cipher_encrypt_one(tfm, foc->val, (void *)buf);
foc->len = TCP_FASTOPEN_COOKIE_SIZE;
- ok = true;
+ return true;
}
- rcu_read_unlock();
- return ok;
+#endif
+ return false;
}
/* Generate the fastopen cookie by doing aes128 encryption on both
@@ -138,37 +179,20 @@ static bool __tcp_fastopen_cookie_gen(struct sock *sk, const void *path,
*
* XXX (TFO) - refactor when TCP_FASTOPEN_COOKIE_SIZE != AES_BLOCK_SIZE.
*/
-static bool tcp_fastopen_cookie_gen(struct sock *sk,
+static void tcp_fastopen_cookie_gen(struct sock *sk,
struct request_sock *req,
struct sk_buff *syn,
struct tcp_fastopen_cookie *foc)
{
- if (req->rsk_ops->family == AF_INET) {
- const struct iphdr *iph = ip_hdr(syn);
-
- __be32 path[4] = { iph->saddr, iph->daddr, 0, 0 };
- return __tcp_fastopen_cookie_gen(sk, path, foc);
- }
-
-#if IS_ENABLED(CONFIG_IPV6)
- if (req->rsk_ops->family == AF_INET6) {
- const struct ipv6hdr *ip6h = ipv6_hdr(syn);
- struct tcp_fastopen_cookie tmp;
-
- if (__tcp_fastopen_cookie_gen(sk, &ip6h->saddr, &tmp)) {
- struct in6_addr *buf = &tmp.addr;
- int i;
+ struct tcp_fastopen_context *ctx;
- for (i = 0; i < 4; i++)
- buf->s6_addr32[i] ^= ip6h->daddr.s6_addr32[i];
- return __tcp_fastopen_cookie_gen(sk, buf, foc);
- }
- }
-#endif
- return false;
+ rcu_read_lock();
+ ctx = tcp_fastopen_get_ctx(sk);
+ if (ctx)
+ __tcp_fastopen_cookie_gen_cipher(req, syn, ctx->tfm[0], foc);
+ rcu_read_unlock();
}
-
/* If an incoming SYN or SYNACK frame contains a payload and/or FIN,
* queue this additional data / FIN.
*/
@@ -212,6 +236,35 @@ void tcp_fastopen_add_skb(struct sock *sk, struct sk_buff *skb)
tcp_fin(sk);
}
+/* returns 0 - no key match, 1 for primary, 2 for backup */
+static int tcp_fastopen_cookie_gen_check(struct sock *sk,
+ struct request_sock *req,
+ struct sk_buff *syn,
+ struct tcp_fastopen_cookie *orig,
+ struct tcp_fastopen_cookie *valid_foc)
+{
+ struct tcp_fastopen_cookie search_foc = { .len = -1 };
+ struct tcp_fastopen_cookie *foc = valid_foc;
+ struct tcp_fastopen_context *ctx;
+ int i, ret = 0;
+
+ rcu_read_lock();
+ ctx = tcp_fastopen_get_ctx(sk);
+ if (!ctx)
+ goto out;
+ for (i = 0; i < tcp_fastopen_context_len(ctx); i++) {
+ __tcp_fastopen_cookie_gen_cipher(req, syn, ctx->tfm[i], foc);
+ if (tcp_fastopen_cookie_match(foc, orig)) {
+ ret = i + 1;
+ goto out;
+ }
+ foc = &search_foc;
+ }
+out:
+ rcu_read_unlock();
+ return ret;
+}
+
static struct sock *tcp_fastopen_create_child(struct sock *sk,
struct sk_buff *skb,
struct request_sock *req)
@@ -331,6 +384,7 @@ struct sock *tcp_try_fastopen(struct sock *sk, struct sk_buff *skb,
int tcp_fastopen = sock_net(sk)->ipv4.sysctl_tcp_fastopen;
struct tcp_fastopen_cookie valid_foc = { .len = -1 };
struct sock *child;
+ int ret = 0;
if (foc->len == 0) /* Client requests a cookie */
NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPFASTOPENCOOKIEREQD);
@@ -346,31 +400,44 @@ struct sock *tcp_try_fastopen(struct sock *sk, struct sk_buff *skb,
tcp_fastopen_no_cookie(sk, dst, TFO_SERVER_COOKIE_NOT_REQD))
goto fastopen;
- if (foc->len >= 0 && /* Client presents or requests a cookie */
- tcp_fastopen_cookie_gen(sk, req, skb, &valid_foc) &&
- foc->len == TCP_FASTOPEN_COOKIE_SIZE &&
- foc->len == valid_foc.len &&
- !memcmp(foc->val, valid_foc.val, foc->len)) {
- /* Cookie is valid. Create a (full) child socket to accept
- * the data in SYN before returning a SYN-ACK to ack the
- * data. If we fail to create the socket, fall back and
- * ack the ISN only but includes the same cookie.
- *
- * Note: Data-less SYN with valid cookie is allowed to send
- * data in SYN_RECV state.
- */
+ if (foc->len == 0) {
+ /* Client requests a cookie. */
+ tcp_fastopen_cookie_gen(sk, req, skb, &valid_foc);
+ } else if (foc->len > 0) {
+ ret = tcp_fastopen_cookie_gen_check(sk, req, skb, foc,
+ &valid_foc);
+ if (!ret) {
+ NET_INC_STATS(sock_net(sk),
+ LINUX_MIB_TCPFASTOPENPASSIVEFAIL);
+ } else {
+ /* Cookie is valid. Create a (full) child socket to
+ * accept the data in SYN before returning a SYN-ACK to
+ * ack the data. If we fail to create the socket, fall
+ * back and ack the ISN only but includes the same
+ * cookie.
+ *
+ * Note: Data-less SYN with valid cookie is allowed to
+ * send data in SYN_RECV state.
+ */
fastopen:
- child = tcp_fastopen_create_child(sk, skb, req);
- if (child) {
- foc->len = -1;
+ child = tcp_fastopen_create_child(sk, skb, req);
+ if (child) {
+ if (ret == 2) {
+ valid_foc.exp = foc->exp;
+ *foc = valid_foc;
+ NET_INC_STATS(sock_net(sk),
+ LINUX_MIB_TCPFASTOPENPASSIVEALTKEY);
+ } else {
+ foc->len = -1;
+ }
+ NET_INC_STATS(sock_net(sk),
+ LINUX_MIB_TCPFASTOPENPASSIVE);
+ return child;
+ }
NET_INC_STATS(sock_net(sk),
- LINUX_MIB_TCPFASTOPENPASSIVE);
- return child;
+ LINUX_MIB_TCPFASTOPENPASSIVEFAIL);
}
- NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPFASTOPENPASSIVEFAIL);
- } else if (foc->len > 0) /* Client presents an invalid cookie */
- NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPFASTOPENPASSIVEFAIL);
-
+ }
valid_foc.exp = foc->exp;
*foc = valid_foc;
return NULL;
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index c61edd023b35..08a477e74cf3 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -3791,6 +3791,8 @@ void tcp_parse_options(const struct net *net,
length--;
continue;
default:
+ if (length < 2)
+ return;
opsize = *ptr++;
if (opsize < 2) /* "silly options" */
return;
generated by cgit v1.2.3-55-g7522 (git 2.39.0) at 2025-04-26 19:00:07 +0200