1 files changed, 247 insertions, 58 deletions

diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 170d6e7f31d3..3d2fe7712ac5 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -14,6 +14,7 @@
  *
  */
 
+#include <linux/compat.h>
 #include <linux/workqueue.h>
 #include <net/xfrm.h>
 #include <linux/pfkeyv2.h>
@@ -44,7 +45,6 @@ static void xfrm_state_gc_task(struct work_struct *work);
  */
 
 static unsigned int xfrm_state_hashmax __read_mostly = 1 * 1024 * 1024;
-static __read_mostly seqcount_t xfrm_state_hash_generation = SEQCNT_ZERO(xfrm_state_hash_generation);
 static struct kmem_cache *xfrm_state_cache __ro_after_init;
 
 static DECLARE_WORK(xfrm_state_gc_work, xfrm_state_gc_task);
@@ -79,10 +79,16 @@ xfrm_spi_hash(struct net *net, const xfrm_address_t *daddr,
 	return __xfrm_spi_hash(daddr, spi, proto, family, net->xfrm.state_hmask);
 }
 
+static unsigned int xfrm_seq_hash(struct net *net, u32 seq)
+{
+	return __xfrm_seq_hash(seq, net->xfrm.state_hmask);
+}
+
 static void xfrm_hash_transfer(struct hlist_head *list,
 			       struct hlist_head *ndsttable,
 			       struct hlist_head *nsrctable,
 			       struct hlist_head *nspitable,
+			       struct hlist_head *nseqtable,
 			       unsigned int nhashmask)
 {
 	struct hlist_node *tmp;
@@ -107,6 +113,11 @@ static void xfrm_hash_transfer(struct hlist_head *list,
 					    nhashmask);
 			hlist_add_head_rcu(&x->byspi, nspitable + h);
 		}
+
+		if (x->km.seq) {
+			h = __xfrm_seq_hash(x->km.seq, nhashmask);
+			hlist_add_head_rcu(&x->byseq, nseqtable + h);
+		}
 	}
 }
 
@@ -118,7 +129,7 @@ static unsigned long xfrm_hash_new_size(unsigned int state_hmask)
 static void xfrm_hash_resize(struct work_struct *work)
 {
 	struct net *net = container_of(work, struct net, xfrm.state_hash_work);
-	struct hlist_head *ndst, *nsrc, *nspi, *odst, *osrc, *ospi;
+	struct hlist_head *ndst, *nsrc, *nspi, *nseq, *odst, *osrc, *ospi, *oseq;
 	unsigned long nsize, osize;
 	unsigned int nhashmask, ohashmask;
 	int i;
@@ -138,25 +149,34 @@ static void xfrm_hash_resize(struct work_struct *work)
 		xfrm_hash_free(nsrc, nsize);
 		return;
 	}
+	nseq = xfrm_hash_alloc(nsize);
+	if (!nseq) {
+		xfrm_hash_free(ndst, nsize);
+		xfrm_hash_free(nsrc, nsize);
+		xfrm_hash_free(nspi, nsize);
+		return;
+	}
 
 	spin_lock_bh(&net->xfrm.xfrm_state_lock);
-	write_seqcount_begin(&xfrm_state_hash_generation);
+	write_seqcount_begin(&net->xfrm.xfrm_state_hash_generation);
 
 	nhashmask = (nsize / sizeof(struct hlist_head)) - 1U;
 	odst = xfrm_state_deref_prot(net->xfrm.state_bydst, net);
 	for (i = net->xfrm.state_hmask; i >= 0; i--)
-		xfrm_hash_transfer(odst + i, ndst, nsrc, nspi, nhashmask);
+		xfrm_hash_transfer(odst + i, ndst, nsrc, nspi, nseq, nhashmask);
 
 	osrc = xfrm_state_deref_prot(net->xfrm.state_bysrc, net);
 	ospi = xfrm_state_deref_prot(net->xfrm.state_byspi, net);
+	oseq = xfrm_state_deref_prot(net->xfrm.state_byseq, net);
 	ohashmask = net->xfrm.state_hmask;
 
 	rcu_assign_pointer(net->xfrm.state_bydst, ndst);
 	rcu_assign_pointer(net->xfrm.state_bysrc, nsrc);
 	rcu_assign_pointer(net->xfrm.state_byspi, nspi);
+	rcu_assign_pointer(net->xfrm.state_byseq, nseq);
 	net->xfrm.state_hmask = nhashmask;
 
-	write_seqcount_end(&xfrm_state_hash_generation);
+	write_seqcount_end(&net->xfrm.xfrm_state_hash_generation);
 	spin_unlock_bh(&net->xfrm.xfrm_state_lock);
 
 	osize = (ohashmask + 1) * sizeof(struct hlist_head);
@@ -166,6 +186,7 @@ static void xfrm_hash_resize(struct work_struct *work)
 	xfrm_hash_free(odst, osize);
 	xfrm_hash_free(osrc, osize);
 	xfrm_hash_free(ospi, osize);
+	xfrm_hash_free(oseq, osize);
 }
 
 static DEFINE_SPINLOCK(xfrm_state_afinfo_lock);
@@ -612,7 +633,7 @@ struct xfrm_state *xfrm_state_alloc(struct net *net)
 {
 	struct xfrm_state *x;
 
-	x = kmem_cache_alloc(xfrm_state_cache, GFP_ATOMIC | __GFP_ZERO);
+	x = kmem_cache_zalloc(xfrm_state_cache, GFP_ATOMIC);
 
 	if (x) {
 		write_pnet(&x->xs_net, net);
@@ -622,6 +643,7 @@ struct xfrm_state *xfrm_state_alloc(struct net *net)
 		INIT_HLIST_NODE(&x->bydst);
 		INIT_HLIST_NODE(&x->bysrc);
 		INIT_HLIST_NODE(&x->byspi);
+		INIT_HLIST_NODE(&x->byseq);
 		hrtimer_init(&x->mtimer, CLOCK_BOOTTIME, HRTIMER_MODE_ABS_SOFT);
 		x->mtimer.function = xfrm_timer_handler;
 		timer_setup(&x->rtimer, xfrm_replay_timer_handler, 0);
@@ -665,6 +687,8 @@ int __xfrm_state_delete(struct xfrm_state *x)
 		list_del(&x->km.all);
 		hlist_del_rcu(&x->bydst);
 		hlist_del_rcu(&x->bysrc);
+		if (x->km.seq)
+			hlist_del_rcu(&x->byseq);
 		if (x->id.spi)
 			hlist_del_rcu(&x->byspi);
 		net->xfrm.state_num--;
@@ -727,7 +751,7 @@ xfrm_dev_state_flush_secctx_check(struct net *net, struct net_device *dev, bool
 
 	for (i = 0; i <= net->xfrm.state_hmask; i++) {
 		struct xfrm_state *x;
-		struct xfrm_state_offload *xso;
+		struct xfrm_dev_offload *xso;
 
 		hlist_for_each_entry(x, net->xfrm.state_bydst+i, bydst) {
 			xso = &x->xso;
@@ -811,7 +835,7 @@ int xfrm_dev_state_flush(struct net *net, struct net_device *dev, bool task_vali
 	err = -ESRCH;
 	for (i = 0; i <= net->xfrm.state_hmask; i++) {
 		struct xfrm_state *x;
-		struct xfrm_state_offload *xso;
+		struct xfrm_dev_offload *xso;
 restart:
 		hlist_for_each_entry(x, net->xfrm.state_bydst+i, bydst) {
 			xso = &x->xso;
@@ -1019,8 +1043,10 @@ static void xfrm_state_look_at(struct xfrm_policy *pol, struct xfrm_state *x,
 	 */
 	if (x->km.state == XFRM_STATE_VALID) {
 		if ((x->sel.family &&
-		     !xfrm_selector_match(&x->sel, fl, x->sel.family)) ||
-		    !security_xfrm_state_pol_flow_match(x, pol, fl))
+		     (x->sel.family != family ||
+		      !xfrm_selector_match(&x->sel, fl, family))) ||
+		    !security_xfrm_state_pol_flow_match(x, pol,
+							&fl->u.__fl_common))
 			return;
 
 		if (!*best ||
@@ -1032,8 +1058,11 @@ static void xfrm_state_look_at(struct xfrm_policy *pol, struct xfrm_state *x,
 		*acq_in_progress = 1;
 	} else if (x->km.state == XFRM_STATE_ERROR ||
 		   x->km.state == XFRM_STATE_EXPIRED) {
-		if (xfrm_selector_match(&x->sel, fl, x->sel.family) &&
-		    security_xfrm_state_pol_flow_match(x, pol, fl))
+		if ((!x->sel.family ||
+		     (x->sel.family == family &&
+		      xfrm_selector_match(&x->sel, fl, family))) &&
+		    security_xfrm_state_pol_flow_match(x, pol,
+						       &fl->u.__fl_common))
 			*error = -ESRCH;
 	}
 }
@@ -1058,7 +1087,7 @@ xfrm_state_find(const xfrm_address_t *daddr, const xfrm_address_t *saddr,
 
 	to_put = NULL;
 
-	sequence = read_seqcount_begin(&xfrm_state_hash_generation);
+	sequence = read_seqcount_begin(&net->xfrm.xfrm_state_hash_generation);
 
 	rcu_read_lock();
 	h = xfrm_dst_hash(net, daddr, saddr, tmpl->reqid, encap_family);
@@ -1072,7 +1101,7 @@ xfrm_state_find(const xfrm_address_t *daddr, const xfrm_address_t *saddr,
 		    tmpl->mode == x->props.mode &&
 		    tmpl->id.proto == x->id.proto &&
 		    (tmpl->id.spi == x->id.spi || !tmpl->id.spi))
-			xfrm_state_look_at(pol, x, fl, encap_family,
+			xfrm_state_look_at(pol, x, fl, family,
 					   &best, &acquire_in_progress, &error);
 	}
 	if (best || acquire_in_progress)
@@ -1089,7 +1118,7 @@ xfrm_state_find(const xfrm_address_t *daddr, const xfrm_address_t *saddr,
 		    tmpl->mode == x->props.mode &&
 		    tmpl->id.proto == x->id.proto &&
 		    (tmpl->id.spi == x->id.spi || !tmpl->id.spi))
-			xfrm_state_look_at(pol, x, fl, encap_family,
+			xfrm_state_look_at(pol, x, fl, family,
 					   &best, &acquire_in_progress, &error);
 	}
 
@@ -1144,6 +1173,10 @@ found:
 				h = xfrm_spi_hash(net, &x->id.daddr, x->id.spi, x->id.proto, encap_family);
 				hlist_add_head_rcu(&x->byspi, net->xfrm.state_byspi + h);
 			}
+			if (x->km.seq) {
+				h = xfrm_seq_hash(net, x->km.seq);
+				hlist_add_head_rcu(&x->byseq, net->xfrm.state_byseq + h);
+			}
 			x->lft.hard_add_expires_seconds = net->xfrm.sysctl_acq_expires;
 			hrtimer_start(&x->mtimer,
 				      ktime_set(net->xfrm.sysctl_acq_expires, 0),
@@ -1171,7 +1204,7 @@ out:
 	if (to_put)
 		xfrm_state_put(to_put);
 
-	if (read_seqcount_retry(&xfrm_state_hash_generation, sequence)) {
+	if (read_seqcount_retry(&net->xfrm.xfrm_state_hash_generation, sequence)) {
 		*err = -EAGAIN;
 		if (x) {
 			xfrm_state_put(x);
@@ -1259,6 +1292,12 @@ static void __xfrm_state_insert(struct xfrm_state *x)
 		hlist_add_head_rcu(&x->byspi, net->xfrm.state_byspi + h);
 	}
 
+	if (x->km.seq) {
+		h = xfrm_seq_hash(net, x->km.seq);
+
+		hlist_add_head_rcu(&x->byseq, net->xfrm.state_byseq + h);
+	}
+
 	hrtimer_start(&x->mtimer, ktime_set(1, 0), HRTIMER_MODE_REL_SOFT);
 	if (x->replay_maxage)
 		mod_timer(&x->rtimer, jiffies + x->replay_maxage);
@@ -1441,6 +1480,30 @@ out:
 EXPORT_SYMBOL(xfrm_state_add);
 
 #ifdef CONFIG_XFRM_MIGRATE
+static inline int clone_security(struct xfrm_state *x, struct xfrm_sec_ctx *security)
+{
+	struct xfrm_user_sec_ctx *uctx;
+	int size = sizeof(*uctx) + security->ctx_len;
+	int err;
+
+	uctx = kmalloc(size, GFP_KERNEL);
+	if (!uctx)
+		return -ENOMEM;
+
+	uctx->exttype = XFRMA_SEC_CTX;
+	uctx->len = size;
+	uctx->ctx_doi = security->ctx_doi;
+	uctx->ctx_alg = security->ctx_alg;
+	uctx->ctx_len = security->ctx_len;
+	memcpy(uctx + 1, security->ctx_str, security->ctx_len);
+	err = security_xfrm_state_alloc(x, uctx);
+	kfree(uctx);
+	if (err)
+		return err;
+
+	return 0;
+}
+
 static struct xfrm_state *xfrm_state_clone(struct xfrm_state *orig,
 					   struct xfrm_encap_tmpl *encap)
 {
@@ -1497,6 +1560,10 @@ static struct xfrm_state *xfrm_state_clone(struct xfrm_state *orig,
 			goto error;
 	}
 
+	if (orig->security)
+		if (clone_security(x, orig->security))
+			goto error;
+
 	if (orig->coaddr) {
 		x->coaddr = kmemdup(orig->coaddr, sizeof(*x->coaddr),
 				    GFP_KERNEL);
@@ -1510,9 +1577,7 @@ static struct xfrm_state *xfrm_state_clone(struct xfrm_state *orig,
 	}
 
 	memcpy(&x->mark, &orig->mark, sizeof(x->mark));
-
-	if (xfrm_init_state(x) < 0)
-		goto error;
+	memcpy(&x->props.smark, &orig->props.smark, sizeof(x->props.smark));
 
 	x->props.flags = orig->props.flags;
 	x->props.extra_flags = orig->props.extra_flags;
@@ -1521,11 +1586,15 @@ static struct xfrm_state *xfrm_state_clone(struct xfrm_state *orig,
 	x->tfcpad = orig->tfcpad;
 	x->replay_maxdiff = orig->replay_maxdiff;
 	x->replay_maxage = orig->replay_maxage;
-	x->curlft.add_time = orig->curlft.add_time;
+	memcpy(&x->curlft, &orig->curlft, sizeof(x->curlft));
 	x->km.state = orig->km.state;
 	x->km.seq = orig->km.seq;
 	x->replay = orig->replay;
 	x->preplay = orig->preplay;
+	x->mapping_maxage = orig->mapping_maxage;
+	x->lastused = orig->lastused;
+	x->new_mapping = 0;
+	x->new_mapping_sport = 0;
 
 	return x;
 
@@ -1535,7 +1604,8 @@ out:
 	return NULL;
 }
 
-struct xfrm_state *xfrm_migrate_state_find(struct xfrm_migrate *m, struct net *net)
+struct xfrm_state *xfrm_migrate_state_find(struct xfrm_migrate *m, struct net *net,
+						u32 if_id)
 {
 	unsigned int h;
 	struct xfrm_state *x = NULL;
@@ -1551,6 +1621,8 @@ struct xfrm_state *xfrm_migrate_state_find(struct xfrm_migrate *m, struct net *n
 				continue;
 			if (m->reqid && x->props.reqid != m->reqid)
 				continue;
+			if (if_id != 0 && x->if_id != if_id)
+				continue;
 			if (!xfrm_addr_equal(&x->id.daddr, &m->old_daddr,
 					     m->old_family) ||
 			    !xfrm_addr_equal(&x->props.saddr, &m->old_saddr,
@@ -1566,6 +1638,8 @@ struct xfrm_state *xfrm_migrate_state_find(struct xfrm_migrate *m, struct net *n
 			if (x->props.mode != m->mode ||
 			    x->id.proto != m->proto)
 				continue;
+			if (if_id != 0 && x->if_id != if_id)
+				continue;
 			if (!xfrm_addr_equal(&x->id.daddr, &m->old_daddr,
 					     m->old_family) ||
 			    !xfrm_addr_equal(&x->props.saddr, &m->old_saddr,
@@ -1592,6 +1666,11 @@ struct xfrm_state *xfrm_state_migrate(struct xfrm_state *x,
 	if (!xc)
 		return NULL;
 
+	xc->props.family = m->new_family;
+
+	if (xfrm_init_state(xc) < 0)
+		goto error;
+
 	memcpy(&xc->id.daddr, &m->new_daddr, sizeof(xc->id.daddr));
 	memcpy(&xc->props.saddr, &m->new_saddr, sizeof(xc->props.saddr));
 
@@ -1899,20 +1978,18 @@ xfrm_state_sort(struct xfrm_state **dst, struct xfrm_state **src, int n,
 
 static struct xfrm_state *__xfrm_find_acq_byseq(struct net *net, u32 mark, u32 seq)
 {
-	int i;
-
-	for (i = 0; i <= net->xfrm.state_hmask; i++) {
-		struct xfrm_state *x;
+	unsigned int h = xfrm_seq_hash(net, seq);
+	struct xfrm_state *x;
 
-		hlist_for_each_entry(x, net->xfrm.state_bydst+i, bydst) {
-			if (x->km.seq == seq &&
-			    (mark & x->mark.m) == x->mark.v &&
-			    x->km.state == XFRM_STATE_ACQ) {
-				xfrm_state_hold(x);
-				return x;
-			}
+	hlist_for_each_entry_rcu(x, net->xfrm.state_byseq + h, byseq) {
+		if (x->km.seq == seq &&
+		    (mark & x->mark.m) == x->mark.v &&
+		    x->km.state == XFRM_STATE_ACQ) {
+			xfrm_state_hold(x);
+			return x;
 		}
 	}
+
 	return NULL;
 }
 
@@ -1972,6 +2049,7 @@ int xfrm_alloc_spi(struct xfrm_state *x, u32 low, u32 high)
 	int err = -ENOENT;
 	__be32 minspi = htonl(low);
 	__be32 maxspi = htonl(high);
+	__be32 newspi = 0;
 	u32 mark = x->mark.v & x->mark.m;
 
 	spin_lock_bh(&x->lock);
@@ -1990,21 +2068,22 @@ int xfrm_alloc_spi(struct xfrm_state *x, u32 low, u32 high)
 			xfrm_state_put(x0);
 			goto unlock;
 		}
-		x->id.spi = minspi;
+		newspi = minspi;
 	} else {
 		u32 spi = 0;
 		for (h = 0; h < high-low+1; h++) {
-			spi = low + prandom_u32()%(high-low+1);
+			spi = low + prandom_u32_max(high - low + 1);
 			x0 = xfrm_state_lookup(net, mark, &x->id.daddr, htonl(spi), x->id.proto, x->props.family);
 			if (x0 == NULL) {
-				x->id.spi = htonl(spi);
+				newspi = htonl(spi);
 				break;
 			}
 			xfrm_state_put(x0);
 		}
 	}
-	if (x->id.spi) {
+	if (newspi) {
 		spin_lock_bh(&net->xfrm.xfrm_state_lock);
+		x->id.spi = newspi;
 		h = xfrm_spi_hash(net, &x->id.daddr, x->id.spi, x->id.proto, x->props.family);
 		hlist_add_head_rcu(&x->byspi, net->xfrm.state_byspi + h);
 		spin_unlock_bh(&net->xfrm.xfrm_state_lock);
@@ -2110,7 +2189,7 @@ static void xfrm_replay_timer_handler(struct timer_list *t)
 
 	if (x->km.state == XFRM_STATE_VALID) {
 		if (xfrm_aevent_is_on(xs_net(x)))
-			x->repl->notify(x, XFRM_REPLAY_TIMEOUT);
+			xfrm_replay_notify(x, XFRM_REPLAY_TIMEOUT);
 		else
 			x->xflags |= XFRM_TIME_DEFER;
 	}
@@ -2175,7 +2254,7 @@ int km_query(struct xfrm_state *x, struct xfrm_tmpl *t, struct xfrm_policy *pol)
 }
 EXPORT_SYMBOL(km_query);
 
-int km_new_mapping(struct xfrm_state *x, xfrm_address_t *ipaddr, __be16 sport)
+static int __km_new_mapping(struct xfrm_state *x, xfrm_address_t *ipaddr, __be16 sport)
 {
 	int err = -EINVAL;
 	struct xfrm_mgr *km;
@@ -2190,6 +2269,24 @@ int km_new_mapping(struct xfrm_state *x, xfrm_address_t *ipaddr, __be16 sport)
 	rcu_read_unlock();
 	return err;
 }
+
+int km_new_mapping(struct xfrm_state *x, xfrm_address_t *ipaddr, __be16 sport)
+{
+	int ret = 0;
+
+	if (x->mapping_maxage) {
+		if ((jiffies / HZ - x->new_mapping) > x->mapping_maxage ||
+		    x->new_mapping_sport != sport) {
+			x->new_mapping_sport = sport;
+			x->new_mapping = jiffies / HZ;
+			ret = __km_new_mapping(x, ipaddr, sport);
+		}
+	} else {
+		ret = __km_new_mapping(x, ipaddr, sport);
+	}
+
+	return ret;
+}
 EXPORT_SYMBOL(km_new_mapping);
 
 void km_policy_expired(struct xfrm_policy *pol, int dir, int hard, u32 portid)
@@ -2264,17 +2361,74 @@ static bool km_is_alive(const struct km_event *c)
 	return is_alive;
 }
 
-int xfrm_user_policy(struct sock *sk, int optname, u8 __user *optval, int optlen)
+#if IS_ENABLED(CONFIG_XFRM_USER_COMPAT)
+static DEFINE_SPINLOCK(xfrm_translator_lock);
+static struct xfrm_translator __rcu *xfrm_translator;
+
+struct xfrm_translator *xfrm_get_translator(void)
+{
+	struct xfrm_translator *xtr;
+
+	rcu_read_lock();
+	xtr = rcu_dereference(xfrm_translator);
+	if (unlikely(!xtr))
+		goto out;
+	if (!try_module_get(xtr->owner))
+		xtr = NULL;
+out:
+	rcu_read_unlock();
+	return xtr;
+}
+EXPORT_SYMBOL_GPL(xfrm_get_translator);
+
+void xfrm_put_translator(struct xfrm_translator *xtr)
+{
+	module_put(xtr->owner);
+}
+EXPORT_SYMBOL_GPL(xfrm_put_translator);
+
+int xfrm_register_translator(struct xfrm_translator *xtr)
+{
+	int err = 0;
+
+	spin_lock_bh(&xfrm_translator_lock);
+	if (unlikely(xfrm_translator != NULL))
+		err = -EEXIST;
+	else
+		rcu_assign_pointer(xfrm_translator, xtr);
+	spin_unlock_bh(&xfrm_translator_lock);
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(xfrm_register_translator);
+
+int xfrm_unregister_translator(struct xfrm_translator *xtr)
+{
+	int err = 0;
+
+	spin_lock_bh(&xfrm_translator_lock);
+	if (likely(xfrm_translator != NULL)) {
+		if (rcu_access_pointer(xfrm_translator) != xtr)
+			err = -EINVAL;
+		else
+			RCU_INIT_POINTER(xfrm_translator, NULL);
+	}
+	spin_unlock_bh(&xfrm_translator_lock);
+	synchronize_rcu();
+
+	return err;
+}
+EXPORT_SYMBOL_GPL(xfrm_unregister_translator);
+#endif
+
+int xfrm_user_policy(struct sock *sk, int optname, sockptr_t optval, int optlen)
 {
 	int err;
 	u8 *data;
 	struct xfrm_mgr *km;
 	struct xfrm_policy *pol = NULL;
 
-	if (in_compat_syscall())
-		return -EOPNOTSUPP;
-
-	if (!optval && !optlen) {
+	if (sockptr_is_null(optval) && !optlen) {
 		xfrm_sk_policy_insert(sk, XFRM_POLICY_IN, NULL);
 		xfrm_sk_policy_insert(sk, XFRM_POLICY_OUT, NULL);
 		__sk_dst_reset(sk);
@@ -2284,10 +2438,26 @@ int xfrm_user_policy(struct sock *sk, int optname, u8 __user *optval, int optlen
 	if (optlen <= 0 || optlen > PAGE_SIZE)
 		return -EMSGSIZE;
 
-	data = memdup_user(optval, optlen);
+	data = memdup_sockptr(optval, optlen);
 	if (IS_ERR(data))
 		return PTR_ERR(data);
 
+	if (in_compat_syscall()) {
+		struct xfrm_translator *xtr = xfrm_get_translator();
+
+		if (!xtr) {
+			kfree(data);
+			return -EOPNOTSUPP;
+		}
+
+		err = xtr->xlate_user_policy_sockptr(&data, optlen);
+		xfrm_put_translator(xtr);
+		if (err) {
+			kfree(data);
+			return err;
+		}
+	}
+
 	err = -EINVAL;
 	rcu_read_lock();
 	list_for_each_entry_rcu(km, &xfrm_km_list, list) {
@@ -2312,22 +2482,20 @@ EXPORT_SYMBOL(xfrm_user_policy);
 
 static DEFINE_SPINLOCK(xfrm_km_lock);
 
-int xfrm_register_km(struct xfrm_mgr *km)
+void xfrm_register_km(struct xfrm_mgr *km)
 {
 	spin_lock_bh(&xfrm_km_lock);
 	list_add_tail_rcu(&km->list, &xfrm_km_list);
 	spin_unlock_bh(&xfrm_km_lock);
-	return 0;
 }
 EXPORT_SYMBOL(xfrm_register_km);
 
-int xfrm_unregister_km(struct xfrm_mgr *km)
+void xfrm_unregister_km(struct xfrm_mgr *km)
 {
 	spin_lock_bh(&xfrm_km_lock);
 	list_del_rcu(&km->list);
 	spin_unlock_bh(&xfrm_km_lock);
 	synchronize_rcu();
-	return 0;
 }
 EXPORT_SYMBOL(xfrm_unregister_km);
 
@@ -2443,7 +2611,8 @@ u32 xfrm_state_mtu(struct xfrm_state *x, int mtu)
 }
 EXPORT_SYMBOL_GPL(xfrm_state_mtu);
 
-int __xfrm_init_state(struct xfrm_state *x, bool init_replay, bool offload)
+int __xfrm_init_state(struct xfrm_state *x, bool init_replay, bool offload,
+		      struct netlink_ext_ack *extack)
 {
 	const struct xfrm_mode *inner_mode;
 	const struct xfrm_mode *outer_mode;
@@ -2451,19 +2620,23 @@ int __xfrm_init_state(struct xfrm_state *x, bool init_replay, bool offload)
 	int err;
 
 	if (family == AF_INET &&
-	    xs_net(x)->ipv4.sysctl_ip_no_pmtu_disc)
+	    READ_ONCE(xs_net(x)->ipv4.sysctl_ip_no_pmtu_disc))
 		x->props.flags |= XFRM_STATE_NOPMTUDISC;
 
 	err = -EPROTONOSUPPORT;
 
 	if (x->sel.family != AF_UNSPEC) {
 		inner_mode = xfrm_get_mode(x->props.mode, x->sel.family);
-		if (inner_mode == NULL)
+		if (inner_mode == NULL) {
+			NL_SET_ERR_MSG(extack, "Requested mode not found");
 			goto error;
+		}
 
 		if (!(inner_mode->flags & XFRM_MODE_FLAG_TUNNEL) &&
-		    family != x->sel.family)
+		    family != x->sel.family) {
+			NL_SET_ERR_MSG(extack, "Only tunnel modes can accommodate a change of family");
 			goto error;
+		}
 
 		x->inner_mode = *inner_mode;
 	} else {
@@ -2471,11 +2644,15 @@ int __xfrm_init_state(struct xfrm_state *x, bool init_replay, bool offload)
 		int iafamily = AF_INET;
 
 		inner_mode = xfrm_get_mode(x->props.mode, x->props.family);
-		if (inner_mode == NULL)
+		if (inner_mode == NULL) {
+			NL_SET_ERR_MSG(extack, "Requested mode not found");
 			goto error;
+		}
 
-		if (!(inner_mode->flags & XFRM_MODE_FLAG_TUNNEL))
+		if (!(inner_mode->flags & XFRM_MODE_FLAG_TUNNEL)) {
+			NL_SET_ERR_MSG(extack, "Only tunnel modes can accommodate an AF_UNSPEC selector");
 			goto error;
+		}
 
 		x->inner_mode = *inner_mode;
 
@@ -2490,24 +2667,27 @@ int __xfrm_init_state(struct xfrm_state *x, bool init_replay, bool offload)
 	}
 
 	x->type = xfrm_get_type(x->id.proto, family);
-	if (x->type == NULL)
+	if (x->type == NULL) {
+		NL_SET_ERR_MSG(extack, "Requested type not found");
 		goto error;
+	}
 
 	x->type_offload = xfrm_get_type_offload(x->id.proto, family, offload);
 
-	err = x->type->init_state(x);
+	err = x->type->init_state(x, extack);
 	if (err)
 		goto error;
 
 	outer_mode = xfrm_get_mode(x->props.mode, family);
 	if (!outer_mode) {
+		NL_SET_ERR_MSG(extack, "Requested mode not found");
 		err = -EPROTONOSUPPORT;
 		goto error;
 	}
 
 	x->outer_mode = *outer_mode;
 	if (init_replay) {
-		err = xfrm_init_replay(x);
+		err = xfrm_init_replay(x, extack);
 		if (err)
 			goto error;
 	}
@@ -2522,7 +2702,7 @@ int xfrm_init_state(struct xfrm_state *x)
 {
 	int err;
 
-	err = __xfrm_init_state(x, true, false);
+	err = __xfrm_init_state(x, true, false, NULL);
 	if (!err)
 		x->km.state = XFRM_STATE_VALID;
 
@@ -2552,13 +2732,20 @@ int __net_init xfrm_state_init(struct net *net)
 	net->xfrm.state_byspi = xfrm_hash_alloc(sz);
 	if (!net->xfrm.state_byspi)
 		goto out_byspi;
+	net->xfrm.state_byseq = xfrm_hash_alloc(sz);
+	if (!net->xfrm.state_byseq)
+		goto out_byseq;
 	net->xfrm.state_hmask = ((sz / sizeof(struct hlist_head)) - 1);
 
 	net->xfrm.state_num = 0;
 	INIT_WORK(&net->xfrm.state_hash_work, xfrm_hash_resize);
 	spin_lock_init(&net->xfrm.xfrm_state_lock);
+	seqcount_spinlock_init(&net->xfrm.xfrm_state_hash_generation,
+			       &net->xfrm.xfrm_state_lock);
 	return 0;
 
+out_byseq:
+	xfrm_hash_free(net->xfrm.state_byspi, sz);
 out_byspi:
 	xfrm_hash_free(net->xfrm.state_bysrc, sz);
 out_bysrc:
@@ -2578,6 +2765,8 @@ void xfrm_state_fini(struct net *net)
 	WARN_ON(!list_empty(&net->xfrm.state_all));
 
 	sz = (net->xfrm.state_hmask + 1) * sizeof(struct hlist_head);
+	WARN_ON(!hlist_empty(net->xfrm.state_byseq));
+	xfrm_hash_free(net->xfrm.state_byseq, sz);
 	WARN_ON(!hlist_empty(net->xfrm.state_byspi));
 	xfrm_hash_free(net->xfrm.state_byspi, sz);
 	WARN_ON(!hlist_empty(net->xfrm.state_bysrc));