--- zzzz-none-000/linux-4.4.271/net/xfrm/xfrm_policy.c	2021-06-03 06:22:09.000000000 +0000
+++ dakota-7530ac-750/linux-4.4.271/net/xfrm/xfrm_policy.c	2023-01-11 09:25:44.000000000 +0000
@@ -44,6 +44,9 @@
 	u8 flags;
 };
 
+static DEFINE_SPINLOCK(xfrm_if_cb_lock);
+static struct xfrm_if_cb const __rcu *xfrm_if_cb __read_mostly;
+
 static DEFINE_SPINLOCK(xfrm_policy_afinfo_lock);
 static struct xfrm_policy_afinfo __rcu *xfrm_policy_afinfo[NPROTO]
 						__read_mostly;
@@ -115,6 +118,12 @@
 	rcu_read_unlock();
 }
 
+/* Called with rcu_read_lock(). */
+static const struct xfrm_if_cb *xfrm_if_get_cb(void)
+{
+	return rcu_dereference(xfrm_if_cb);
+}
+
 static inline struct dst_entry *__xfrm_dst_lookup(struct net *net,
 						  int tos, int oif,
 						  const xfrm_address_t *saddr,
@@ -761,6 +770,7 @@
 	newpos = NULL;
 	hlist_for_each_entry(pol, chain, bydst) {
 		if (pol->type == policy->type &&
+		    pol->if_id == policy->if_id &&
 		    !selector_cmp(&pol->selector, &policy->selector) &&
 		    xfrm_policy_mark_match(policy, pol) &&
 		    xfrm_sec_ctx_match(pol->security, policy->security) &&
@@ -813,8 +823,9 @@
 }
 EXPORT_SYMBOL(xfrm_policy_insert);
 
-struct xfrm_policy *xfrm_policy_bysel_ctx(struct net *net, u32 mark, u8 type,
-					  int dir, struct xfrm_selector *sel,
+struct xfrm_policy *xfrm_policy_bysel_ctx(struct net *net, u32 mark, u32 if_id,
+					  u8 type, int dir,
+					  struct xfrm_selector *sel,
 					  struct xfrm_sec_ctx *ctx, int delete,
 					  int *err)
 {
@@ -827,6 +838,7 @@
 	ret = NULL;
 	hlist_for_each_entry(pol, chain, bydst) {
 		if (pol->type == type &&
+		    pol->if_id == if_id &&
 		    (mark & pol->mark.m) == pol->mark.v &&
 		    !selector_cmp(sel, &pol->selector) &&
 		    xfrm_sec_ctx_match(ctx, pol->security)) {
@@ -852,8 +864,9 @@
 }
 EXPORT_SYMBOL(xfrm_policy_bysel_ctx);
 
-struct xfrm_policy *xfrm_policy_byid(struct net *net, u32 mark, u8 type,
-				     int dir, u32 id, int delete, int *err)
+struct xfrm_policy *xfrm_policy_byid(struct net *net, u32 mark, u32 if_id,
+				     u8 type, int dir, u32 id, int delete,
+				     int *err)
 {
 	struct xfrm_policy *pol, *ret;
 	struct hlist_head *chain;
@@ -868,6 +881,7 @@
 	ret = NULL;
 	hlist_for_each_entry(pol, chain, byidx) {
 		if (pol->type == type && pol->index == id &&
+		    pol->if_id == if_id &&
 		    (mark & pol->mark.m) == pol->mark.v) {
 			xfrm_pol_hold(pol);
 			if (delete) {
@@ -1078,6 +1092,7 @@
 	bool match;
 
 	if (pol->family != family ||
+	    pol->if_id != fl->flowi_xfrm.if_id ||
 	    (fl->flowi_mark & pol->mark.m) != pol->mark.v ||
 	    pol->type != type)
 		return ret;
@@ -1237,7 +1252,8 @@
 
 		match = xfrm_selector_match(&pol->selector, fl, family);
 		if (match) {
-			if ((sk->sk_mark & pol->mark.m) != pol->mark.v) {
+			if ((sk->sk_mark & pol->mark.m) != pol->mark.v ||
+			    pol->if_id != fl->flowi_xfrm.if_id) {
 				pol = NULL;
 				goto out;
 			}
@@ -1364,6 +1380,7 @@
 		newp->lft = old->lft;
 		newp->curlft = old->curlft;
 		newp->mark = old->mark;
+		newp->if_id = old->if_id;
 		newp->action = old->action;
 		newp->flags = old->flags;
 		newp->xfrm_nr = old->xfrm_nr;
@@ -1574,7 +1591,9 @@
 	struct xfrm_dst *xdst = container_of(flo, struct xfrm_dst, flo);
 	struct dst_entry *dst = &xdst->u.dst;
 
-	dst_free(dst);
+	/* Mark DST_OBSOLETE_DEAD to fail the next xfrm_dst_check() */
+	dst->obsolete = DST_OBSOLETE_DEAD;
+	dst_release_immediate(dst);
 }
 
 static const struct flow_cache_ops xfrm_bundle_fc_ops = {
@@ -1604,7 +1623,7 @@
 	default:
 		BUG();
 	}
-	xdst = dst_alloc(dst_ops, NULL, 0, DST_OBSOLETE_NONE, 0);
+	xdst = dst_alloc(dst_ops, NULL, 1, DST_OBSOLETE_NONE, DST_NOGC);
 
 	if (likely(xdst)) {
 		struct dst_entry *dst = &xdst->u.dst;
@@ -1710,10 +1729,11 @@
 
 		if (!dst_prev)
 			dst0 = dst1;
-		else {
-			dst_prev->child = dst_clone(dst1);
-			dst1->flags |= DST_NOHASH;
-		}
+		else
+			/* Ref count is taken during xfrm_alloc_dst()
+			 * No need to do dst_clone() on dst1
+			 */
+			dst_prev->child = dst1;
 
 		xdst->route = dst;
 		dst_copy_metrics(dst1, dst);
@@ -1779,7 +1799,7 @@
 		xfrm_state_put(xfrm[i]);
 free_dst:
 	if (dst0)
-		dst_free(dst0);
+		dst_release_immediate(dst0);
 	dst0 = ERR_PTR(err);
 	goto out;
 }
@@ -1835,8 +1855,8 @@
 {
 	struct net *net = xp_net(pols[0]);
 	struct xfrm_state *xfrm[XFRM_MAX_DEPTH];
-	struct dst_entry *dst;
 	struct xfrm_dst *xdst;
+	struct dst_entry *dst;
 	int err;
 
 	/* Try to instantiate a bundle */
@@ -1924,7 +1944,7 @@
 			continue;
 		}
 
-		nf_reset(skb);
+		nf_reset_no_generic_ct(skb);
 		skb_dst_drop(skb);
 		skb_dst_set(skb, dst);
 
@@ -2063,7 +2083,11 @@
 			pol_dead |= pols[i]->walk.dead;
 		}
 		if (pol_dead) {
-			dst_free(&xdst->u.dst);
+			/* Mark DST_OBSOLETE_DEAD to fail the next
+			 * xfrm_dst_check()
+			 */
+			xdst->u.dst.obsolete = DST_OBSOLETE_DEAD;
+			dst_release_immediate(&xdst->u.dst);
 			xdst = NULL;
 			num_pols = 0;
 			num_xfrms = 0;
@@ -2091,6 +2115,11 @@
 						  xflo->dst_orig);
 	if (IS_ERR(new_xdst)) {
 		err = PTR_ERR(new_xdst);
+		if (err == -EREMOTE) {
+			xfrm_pols_put(pols, num_pols);
+			return NULL;
+		}
+
 		if (err != -EAGAIN)
 			goto error;
 		if (oldflo == NULL)
@@ -2110,11 +2139,12 @@
 	if (xdst) {
 		/* The policies were stolen for newly generated bundle */
 		xdst->num_pols = 0;
-		dst_free(&xdst->u.dst);
+		/* Mark DST_OBSOLETE_DEAD to fail the next xfrm_dst_check() */
+		xdst->u.dst.obsolete = DST_OBSOLETE_DEAD;
+		dst_release_immediate(&xdst->u.dst);
 	}
 
-	/* Flow cache does not have reference, it dst_free()'s,
-	 * but we do need to return one reference for original caller */
+	/* We do need to return one reference for original caller */
 	dst_hold(&new_xdst->u.dst);
 	return &new_xdst->flo;
 
@@ -2137,9 +2167,11 @@
 inc_error:
 	XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTPOLERROR);
 error:
-	if (xdst != NULL)
-		dst_free(&xdst->u.dst);
-	else
+	if (xdst != NULL) {
+		/* Mark DST_OBSOLETE_DEAD to fail the next xfrm_dst_check() */
+		xdst->u.dst.obsolete = DST_OBSOLETE_DEAD;
+		dst_release_immediate(&xdst->u.dst);
+	} else
 		xfrm_pols_put(pols, num_pols);
 	return ERR_PTR(err);
 }
@@ -2203,6 +2235,9 @@
 			if (IS_ERR(xdst)) {
 				xfrm_pols_put(pols, num_pols);
 				err = PTR_ERR(xdst);
+				if (err == -EREMOTE)
+					goto nopol;
+
 				goto dropdst;
 			} else if (xdst == NULL) {
 				num_xfrms = 0;
@@ -2402,12 +2437,20 @@
 			  unsigned int family, int reverse)
 {
 	struct xfrm_policy_afinfo *afinfo = xfrm_policy_get_afinfo(family);
+	const struct xfrm_if_cb *ifcb = xfrm_if_get_cb();
+	struct xfrm_if *xi;
 	int err;
 
 	if (unlikely(afinfo == NULL))
 		return -EAFNOSUPPORT;
 
 	afinfo->decode_session(skb, fl, reverse);
+	if (ifcb) {
+		xi = ifcb->decode_session(skb);
+		if (xi)
+			fl->flowi_xfrm.if_id = xi->p.if_id;
+	}
+
 	err = security_xfrm_decode_session(skb, &fl->flowi_secid);
 	xfrm_policy_put_afinfo(afinfo);
 	return err;
@@ -2629,10 +2672,12 @@
 	 * notice.  That's what we are validating here via the
 	 * stale_bundle() check.
 	 *
-	 * When a policy's bundle is pruned, we dst_free() the XFRM
-	 * dst which causes it's ->obsolete field to be set to
-	 * DST_OBSOLETE_DEAD.  If an XFRM dst has been pruned like
-	 * this, we want to force a new route lookup.
+	 * When an xdst is removed from flow cache, DST_OBSOLETE_DEAD will
+	 * be marked on it.
+	 * When a dst is removed from the fib tree, DST_OBSOLETE_DEAD will
+	 * be marked on it.
+	 * Both will force stable_bundle() to fail on any xdst bundle with
+	 * this dst linked in it.
 	 */
 	if (dst->obsolete < 0 && !stale_bundle(dst))
 		return dst;
@@ -2873,6 +2918,21 @@
 	.notifier_call	= xfrm_dev_event,
 };
 
+void xfrm_if_register_cb(const struct xfrm_if_cb *ifcb)
+{
+	spin_lock(&xfrm_if_cb_lock);
+	rcu_assign_pointer(xfrm_if_cb, ifcb);
+	spin_unlock(&xfrm_if_cb_lock);
+}
+EXPORT_SYMBOL(xfrm_if_register_cb);
+
+void xfrm_if_unregister_cb(void)
+{
+	RCU_INIT_POINTER(xfrm_if_cb, NULL);
+	synchronize_rcu();
+}
+EXPORT_SYMBOL(xfrm_if_unregister_cb);
+
 #ifdef CONFIG_XFRM_STATISTICS
 static int __net_init xfrm_statistics_init(struct net *net)
 {
@@ -3001,6 +3061,9 @@
 	spin_lock_init(&net->xfrm.xfrm_state_lock);
 	rwlock_init(&net->xfrm.xfrm_policy_lock);
 	mutex_init(&net->xfrm.xfrm_cfg_mutex);
+	spin_lock_init(&net->xfrm.xfrm_event_lock);
+
+	INIT_LIST_HEAD(&net->xfrm.event_notifier_list);
 
 	rv = xfrm_statistics_init(net);
 	if (rv < 0)
@@ -3050,6 +3113,10 @@
 {
 	register_pernet_subsys(&xfrm_net_ops);
 	xfrm_input_init();
+	xfrm_output_init();
+
+	RCU_INIT_POINTER(xfrm_if_cb, NULL);
+	synchronize_rcu();
 }
 
 #ifdef CONFIG_AUDITSYSCALL