Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

Move the bpf verifier trace check into the new switch statement in
HEAD.

Resolve the overlapping changes in hinic, where bug fixes overlap
the addition of VF support.

Signed-off-by: David S. Miller <davem@davemloft.net>

27 files changed, 237 insertions, 81 deletions

diff --git a/net/core/dev.c b/net/core/dev.c
index f937a3ff668d..f36bd3b21997 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -9006,11 +9006,13 @@ static void netdev_sync_lower_features(struct net_device *upper,
 			netdev_dbg(upper, "Disabling feature %pNF on lower dev %s.\n",
 				   &feature, lower->name);
 			lower->wanted_features &= ~feature;
-			netdev_update_features(lower);
+			__netdev_update_features(lower);
 
 			if (unlikely(lower->features & feature))
 				netdev_WARN(upper, "failed to disable %pNF on %s!\n",
 					    &feature, lower->name);
+			else
+				netdev_features_change(lower);
 		}
 	}
 }
diff --git a/net/core/filter.c b/net/core/filter.c
index f8a3c7e9d027..822d662f97ef 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -2579,8 +2579,8 @@ BPF_CALL_4(bpf_msg_pop_data, struct sk_msg *, msg, u32, start,
 			}
 			pop = 0;
 		} else if (pop >= sge->length - a) {
-			sge->length = a;
 			pop -= (sge->length - a);
+			sge->length = a;
 		}
 	}
 
diff --git a/net/core/netprio_cgroup.c b/net/core/netprio_cgroup.c
index 8881dd943dd0..9bd4cab7d510 100644
--- a/net/core/netprio_cgroup.c
+++ b/net/core/netprio_cgroup.c
@@ -236,6 +236,8 @@ static void net_prio_attach(struct cgroup_taskset *tset)
 	struct task_struct *p;
 	struct cgroup_subsys_state *css;
 
+	cgroup_sk_alloc_disable();
+
 	cgroup_taskset_for_each(p, css, tset) {
 		void *v = (void *)(unsigned long)css->id;
 
diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c
index 0bd10a1f477f..a23094b050f8 100644
--- a/net/ipv4/cipso_ipv4.c
+++ b/net/ipv4/cipso_ipv4.c
@@ -1258,7 +1258,8 @@ static int cipso_v4_parsetag_rbm(const struct cipso_v4_doi *doi_def,
 			return ret_val;
 		}
 
-		secattr->flags |= NETLBL_SECATTR_MLS_CAT;
+		if (secattr->attr.mls.cat)
+			secattr->flags |= NETLBL_SECATTR_MLS_CAT;
 	}
 
 	return 0;
@@ -1439,7 +1440,8 @@ static int cipso_v4_parsetag_rng(const struct cipso_v4_doi *doi_def,
 			return ret_val;
 		}
 
-		secattr->flags |= NETLBL_SECATTR_MLS_CAT;
+		if (secattr->attr.mls.cat)
+			secattr->flags |= NETLBL_SECATTR_MLS_CAT;
 	}
 
 	return 0;
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 9cf83cc85e4a..5c218db2dede 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -109,8 +109,10 @@ static void mroute_clean_tables(struct mr_table *mrt, int flags);
 static void ipmr_expire_process(struct timer_list *t);
 
 #ifdef CONFIG_IP_MROUTE_MULTIPLE_TABLES
-#define ipmr_for_each_table(mrt, net) \
-	list_for_each_entry_rcu(mrt, &net->ipv4.mr_tables, list)
+#define ipmr_for_each_table(mrt, net)					\
+	list_for_each_entry_rcu(mrt, &net->ipv4.mr_tables, list,	\
+				lockdep_rtnl_is_held() ||		\
+				list_empty(&net->ipv4.mr_tables))
 
 static struct mr_table *ipmr_mr_table_iter(struct net *net,
 					   struct mr_table *mrt)
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 041f4dcac440..2058b94bfa00 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -915,7 +915,7 @@ void ip_rt_send_redirect(struct sk_buff *skb)
 	/* Check for load limit; set rate_last to the latest sent
 	 * redirect.
 	 */
-	if (peer->rate_tokens == 0 ||
+	if (peer->n_redirects == 0 ||
 	    time_after(jiffies,
 		       (peer->rate_last +
 			(ip_rt_redirect_load << peer->n_redirects)))) {
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 8c1250103959..970064996377 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -476,9 +476,17 @@ static void tcp_tx_timestamp(struct sock *sk, u16 tsflags)
 static inline bool tcp_stream_is_readable(const struct tcp_sock *tp,
 					  int target, struct sock *sk)
 {
-	return (READ_ONCE(tp->rcv_nxt) - READ_ONCE(tp->copied_seq) >= target) ||
-		(sk->sk_prot->stream_memory_read ?
-		sk->sk_prot->stream_memory_read(sk) : false);
+	int avail = READ_ONCE(tp->rcv_nxt) - READ_ONCE(tp->copied_seq);
+
+	if (avail > 0) {
+		if (avail >= target)
+			return true;
+		if (tcp_rmem_pressure(sk))
+			return true;
+	}
+	if (sk->sk_prot->stream_memory_read)
+		return sk->sk_prot->stream_memory_read(sk);
+	return false;
 }
 
 /*
@@ -1756,10 +1764,11 @@ static int tcp_zerocopy_receive(struct sock *sk,
 
 	down_read(&current->mm->mmap_sem);
 
-	ret = -EINVAL;
 	vma = find_vma(current->mm, address);
-	if (!vma || vma->vm_start > address || vma->vm_ops != &tcp_vm_ops)
-		goto out;
+	if (!vma || vma->vm_start > address || vma->vm_ops != &tcp_vm_ops) {
+		up_read(&current->mm->mmap_sem);
+		return -EINVAL;
+	}
 	zc->length = min_t(unsigned long, zc->length, vma->vm_end - address);
 
 	tp = tcp_sk(sk);
@@ -2154,13 +2163,15 @@ skip_copy:
 			tp->urg_data = 0;
 			tcp_fast_path_check(sk);
 		}
-		if (used + offset < skb->len)
-			continue;
 
 		if (TCP_SKB_CB(skb)->has_rxtstamp) {
 			tcp_update_recv_tstamps(skb, &tss);
 			cmsg_flags |= 2;
 		}
+
+		if (used + offset < skb->len)
+			continue;
+
 		if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN)
 			goto found_fin_ok;
 		if (!(flags & MSG_PEEK))
diff --git a/net/ipv4/tcp_bpf.c b/net/ipv4/tcp_bpf.c
index 5a05327f97c1..629aaa9a1eb9 100644
--- a/net/ipv4/tcp_bpf.c
+++ b/net/ipv4/tcp_bpf.c
@@ -125,7 +125,6 @@ static int bpf_tcp_ingress(struct sock *sk, struct sk_psock *psock,
 
 	if (!ret) {
 		msg->sg.start = i;
-		msg->sg.size -= apply_bytes;
 		sk_psock_queue_msg(psock, tmp);
 		sk_psock_data_ready(sk, psock);
 	} else {
@@ -262,14 +261,17 @@ static int tcp_bpf_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
 	struct sk_psock *psock;
 	int copied, ret;
 
+	if (unlikely(flags & MSG_ERRQUEUE))
+		return inet_recv_error(sk, msg, len, addr_len);
+
 	psock = sk_psock_get(sk);
 	if (unlikely(!psock))
 		return tcp_recvmsg(sk, msg, len, nonblock, flags, addr_len);
-	if (unlikely(flags & MSG_ERRQUEUE))
-		return inet_recv_error(sk, msg, len, addr_len);
 	if (!skb_queue_empty(&sk->sk_receive_queue) &&
-	    sk_psock_queue_empty(psock))
+	    sk_psock_queue_empty(psock)) {
+		sk_psock_put(sk, psock);
 		return tcp_recvmsg(sk, msg, len, nonblock, flags, addr_len);
+	}
 	lock_sock(sk);
 msg_bytes_ready:
 	copied = __tcp_bpf_recvmsg(sk, psock, msg, len, flags);
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 7529c2816f2f..ad90102f5dfb 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -4769,7 +4769,8 @@ void tcp_data_ready(struct sock *sk)
 	const struct tcp_sock *tp = tcp_sk(sk);
 	int avail = tp->rcv_nxt - tp->copied_seq;
 
-	if (avail < sk->sk_rcvlowat && !sock_flag(sk, SOCK_DONE))
+	if (avail < sk->sk_rcvlowat && !tcp_rmem_pressure(sk) &&
+	    !sock_flag(sk, SOCK_DONE))
 		return;
 
 	sk->sk_data_ready(sk);
diff --git a/net/ipv6/calipso.c b/net/ipv6/calipso.c
index 221c81f85cbf..8d3f66c310db 100644
--- a/net/ipv6/calipso.c
+++ b/net/ipv6/calipso.c
@@ -1047,7 +1047,8 @@ static int calipso_opt_getattr(const unsigned char *calipso,
 			goto getattr_return;
 		}
 
-		secattr->flags |= NETLBL_SECATTR_MLS_CAT;
+		if (secattr->attr.mls.cat)
+			secattr->flags |= NETLBL_SECATTR_MLS_CAT;
 	}
 
 	secattr->type = NETLBL_NLTYPE_CALIPSO;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index fcf0d5c87d09..a8b4add0b545 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2722,8 +2722,10 @@ static void __ip6_rt_update_pmtu(struct dst_entry *dst, const struct sock *sk,
 	const struct in6_addr *daddr, *saddr;
 	struct rt6_info *rt6 = (struct rt6_info *)dst;
 
-	if (dst_metric_locked(dst, RTAX_MTU))
-		return;
+	/* Note: do *NOT* check dst_metric_locked(dst, RTAX_MTU)
+	 * IPv6 pmtu discovery isn't optional, so 'mtu lock' cannot disable it.
+	 * [see also comment in rt6_mtu_change_route()]
+	 */
 
 	if (iph) {
 		daddr = &iph->daddr;
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index a61e60e94137..1f52a0fa31ed 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -1629,6 +1629,8 @@ bool mptcp_finish_join(struct sock *sk)
 
 	ret = mptcp_pm_allow_new_subflow(msk);
 	if (ret) {
+		subflow->map_seq = msk->ack_seq;
+
 		/* active connections are already on conn_list */
 		spin_lock_bh(&msk->join_list_lock);
 		if (!WARN_ON_ONCE(!list_empty(&subflow->node)))
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 3cf2eeea9d80..53c75b0e5dce 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -1036,6 +1036,16 @@ int mptcp_subflow_create_socket(struct sock *sk, struct socket **new_sock)
 	if (err)
 		return err;
 
+	/* the newly created socket really belongs to the owning MPTCP master
+	 * socket, even if for additional subflows the allocation is performed
+	 * by a kernel workqueue. Adjust inode references, so that the
+	 * procfs/diag interaces really show this one belonging to the correct
+	 * user.
+	 */
+	SOCK_INODE(sf)->i_ino = SOCK_INODE(sk->sk_socket)->i_ino;
+	SOCK_INODE(sf)->i_uid = SOCK_INODE(sk->sk_socket)->i_uid;
+	SOCK_INODE(sf)->i_gid = SOCK_INODE(sk->sk_socket)->i_gid;
+
 	subflow = mptcp_subflow_ctx(sf->sk);
 	pr_debug("subflow=%p", subflow);
 
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index c4582eb71766..1d57b95d3481 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -1519,9 +1519,9 @@ __nf_conntrack_alloc(struct net *net,
 	ct->status = 0;
 	ct->timeout = 0;
 	write_pnet(&ct->ct_net, net);
-	memset(&ct->__nfct_init_offset[0], 0,
+	memset(&ct->__nfct_init_offset, 0,
 	       offsetof(struct nf_conn, proto) -
-	       offsetof(struct nf_conn, __nfct_init_offset[0]));
+	       offsetof(struct nf_conn, __nfct_init_offset));
 
 	nf_ct_zone_add(ct, zone);
 
@@ -2139,8 +2139,19 @@ get_next_corpse(int (*iter)(struct nf_conn *i, void *data),
 		nf_conntrack_lock(lockp);
 		if (*bucket < nf_conntrack_htable_size) {
 			hlist_nulls_for_each_entry(h, n, &nf_conntrack_hash[*bucket], hnnode) {
-				if (NF_CT_DIRECTION(h) != IP_CT_DIR_ORIGINAL)
+				if (NF_CT_DIRECTION(h) != IP_CT_DIR_REPLY)
 					continue;
+				/* All nf_conn objects are added to hash table twice, one
+				 * for original direction tuple, once for the reply tuple.
+				 *
+				 * Exception: In the IPS_NAT_CLASH case, only the reply
+				 * tuple is added (the original tuple already existed for
+				 * a different object).
+				 *
+				 * We only need to call the iterator once for each
+				 * conntrack, so we just use the 'reply' direction
+				 * tuple while iterating.
+				 */
 				ct = nf_ct_tuplehash_to_ctrack(h);
 				if (iter(ct, data))
 					goto found;
diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c
index 4344e572b7f9..42da6e337276 100644
--- a/net/netfilter/nf_flow_table_core.c
+++ b/net/netfilter/nf_flow_table_core.c
@@ -284,7 +284,7 @@ static void flow_offload_del(struct nf_flowtable *flow_table,
 
 	if (nf_flow_has_expired(flow))
 		flow_offload_fixup_ct(flow->ct);
-	else if (test_bit(NF_FLOW_TEARDOWN, &flow->flags))
+	else
 		flow_offload_fixup_ct_timeout(flow->ct);
 
 	flow_offload_free(flow);
@@ -361,8 +361,10 @@ static void nf_flow_offload_gc_step(struct flow_offload *flow, void *data)
 {
 	struct nf_flowtable *flow_table = data;
 
-	if (nf_flow_has_expired(flow) || nf_ct_is_dying(flow->ct) ||
-	    test_bit(NF_FLOW_TEARDOWN, &flow->flags)) {
+	if (nf_flow_has_expired(flow) || nf_ct_is_dying(flow->ct))
+		set_bit(NF_FLOW_TEARDOWN, &flow->flags);
+
+	if (test_bit(NF_FLOW_TEARDOWN, &flow->flags)) {
 		if (test_bit(NF_FLOW_HW, &flow->flags)) {
 			if (!test_bit(NF_FLOW_HW_DYING, &flow->flags))
 				nf_flow_offload_del(flow_table, flow);
diff --git a/net/netfilter/nf_flow_table_offload.c b/net/netfilter/nf_flow_table_offload.c
index a2abb0feab7f..2ff4087007a6 100644
--- a/net/netfilter/nf_flow_table_offload.c
+++ b/net/netfilter/nf_flow_table_offload.c
@@ -820,6 +820,7 @@ static void flow_offload_work_handler(struct work_struct *work)
 			WARN_ON_ONCE(1);
 	}
 
+	clear_bit(NF_FLOW_HW_PENDING, &offload->flow->flags);
 	kfree(offload);
 }
 
@@ -834,9 +835,14 @@ nf_flow_offload_work_alloc(struct nf_flowtable *flowtable,
 {
 	struct flow_offload_work *offload;
 
+	if (test_and_set_bit(NF_FLOW_HW_PENDING, &flow->flags))
+		return NULL;
+
 	offload = kmalloc(sizeof(struct flow_offload_work), GFP_ATOMIC);
-	if (!offload)
+	if (!offload) {
+		clear_bit(NF_FLOW_HW_PENDING, &flow->flags);
 		return NULL;
+	}
 
 	offload->cmd = cmd;
 	offload->flow = flow;
@@ -1059,7 +1065,7 @@ static struct flow_indr_block_entry block_ing_entry = {
 int nf_flow_table_offload_init(void)
 {
 	nf_flow_offload_wq  = alloc_workqueue("nf_flow_table_offload",
-					      WQ_UNBOUND | WQ_MEM_RECLAIM, 0);
+					      WQ_UNBOUND, 0);
 	if (!nf_flow_offload_wq)
 		return -ENOMEM;
 
diff --git a/net/netfilter/nft_set_rbtree.c b/net/netfilter/nft_set_rbtree.c
index 3ffef454d469..62f416bc0579 100644
--- a/net/netfilter/nft_set_rbtree.c
+++ b/net/netfilter/nft_set_rbtree.c
@@ -79,6 +79,10 @@ static bool __nft_rbtree_lookup(const struct net *net, const struct nft_set *set
 				parent = rcu_dereference_raw(parent->rb_left);
 				continue;
 			}
+
+			if (nft_set_elem_expired(&rbe->ext))
+				return false;
+
 			if (nft_rbtree_interval_end(rbe)) {
 				if (nft_set_is_anonymous(set))
 					return false;
@@ -94,6 +98,7 @@ static bool __nft_rbtree_lookup(const struct net *net, const struct nft_set *set
 
 	if (set->flags & NFT_SET_INTERVAL && interval != NULL &&
 	    nft_set_elem_active(&interval->ext, genmask) &&
+	    !nft_set_elem_expired(&interval->ext) &&
 	    nft_rbtree_interval_start(interval)) {
 		*ext = &interval->ext;
 		return true;
@@ -154,6 +159,9 @@ static bool __nft_rbtree_get(const struct net *net, const struct nft_set *set,
 				continue;
 			}
 
+			if (nft_set_elem_expired(&rbe->ext))
+				return false;
+
 			if (!nft_set_ext_exists(&rbe->ext, NFT_SET_EXT_FLAGS) ||
 			    (*nft_set_ext_flags(&rbe->ext) & NFT_SET_ELEM_INTERVAL_END) ==
 			    (flags & NFT_SET_ELEM_INTERVAL_END)) {
@@ -170,6 +178,7 @@ static bool __nft_rbtree_get(const struct net *net, const struct nft_set *set,
 
 	if (set->flags & NFT_SET_INTERVAL && interval != NULL &&
 	    nft_set_elem_active(&interval->ext, genmask) &&
+	    !nft_set_elem_expired(&interval->ext) &&
 	    ((!nft_rbtree_interval_end(interval) &&
 	      !(flags & NFT_SET_ELEM_INTERVAL_END)) ||
 	     (nft_rbtree_interval_end(interval) &&
@@ -418,6 +427,8 @@ static void nft_rbtree_walk(const struct nft_ctx *ctx,
 
 		if (iter->count < iter->skip)
 			goto cont;
+		if (nft_set_elem_expired(&rbe->ext))
+			goto cont;
 		if (!nft_set_elem_active(&rbe->ext, iter->genmask))
 			goto cont;
 
diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c
index 409a3ae47ce2..5e1239cef000 100644
--- a/net/netlabel/netlabel_kapi.c
+++ b/net/netlabel/netlabel_kapi.c
@@ -734,6 +734,12 @@ int netlbl_catmap_getlong(struct netlbl_lsm_catmap *catmap,
 	if ((off & (BITS_PER_LONG - 1)) != 0)
 		return -EINVAL;
 
+	/* a null catmap is equivalent to an empty one */
+	if (!catmap) {
+		*offset = (u32)-1;
+		return 0;
+	}
+
 	if (off < catmap->startbit) {
 		off = catmap->startbit;
 		*offset = off;
diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c
index 25fbd8d9de74..ac5cac0dd24b 100644
--- a/net/sunrpc/auth_gss/auth_gss.c
+++ b/net/sunrpc/auth_gss/auth_gss.c
@@ -2032,7 +2032,6 @@ gss_unwrap_resp_priv(struct rpc_task *task, struct rpc_cred *cred,
 	struct xdr_buf *rcv_buf = &rqstp->rq_rcv_buf;
 	struct kvec *head = rqstp->rq_rcv_buf.head;
 	struct rpc_auth *auth = cred->cr_auth;
-	unsigned int savedlen = rcv_buf->len;
 	u32 offset, opaque_len, maj_stat;
 	__be32 *p;
 
@@ -2043,9 +2042,9 @@ gss_unwrap_resp_priv(struct rpc_task *task, struct rpc_cred *cred,
 	offset = (u8 *)(p) - (u8 *)head->iov_base;
 	if (offset + opaque_len > rcv_buf->len)
 		goto unwrap_failed;
-	rcv_buf->len = offset + opaque_len;
 
-	maj_stat = gss_unwrap(ctx->gc_gss_ctx, offset, rcv_buf);
+	maj_stat = gss_unwrap(ctx->gc_gss_ctx, offset,
+			      offset + opaque_len, rcv_buf);
 	if (maj_stat == GSS_S_CONTEXT_EXPIRED)
 		clear_bit(RPCAUTH_CRED_UPTODATE, &cred->cr_flags);
 	if (maj_stat != GSS_S_COMPLETE)
@@ -2059,10 +2058,9 @@ gss_unwrap_resp_priv(struct rpc_task *task, struct rpc_cred *cred,
 	 */
 	xdr_init_decode(xdr, rcv_buf, p, rqstp);
 
-	auth->au_rslack = auth->au_verfsize + 2 +
-			  XDR_QUADLEN(savedlen - rcv_buf->len);
-	auth->au_ralign = auth->au_verfsize + 2 +
-			  XDR_QUADLEN(savedlen - rcv_buf->len);
+	auth->au_rslack = auth->au_verfsize + 2 + ctx->gc_gss_ctx->slack;
+	auth->au_ralign = auth->au_verfsize + 2 + ctx->gc_gss_ctx->align;
+
 	return 0;
 unwrap_failed:
 	trace_rpcgss_unwrap_failed(task);
diff --git a/net/sunrpc/auth_gss/gss_krb5_crypto.c b/net/sunrpc/auth_gss/gss_krb5_crypto.c
index 6f2d30d7b766..e7180da1fc6a 100644
--- a/net/sunrpc/auth_gss/gss_krb5_crypto.c
+++ b/net/sunrpc/auth_gss/gss_krb5_crypto.c
@@ -851,8 +851,8 @@ out_err:
 }
 
 u32
-gss_krb5_aes_decrypt(struct krb5_ctx *kctx, u32 offset, struct xdr_buf *buf,
-		     u32 *headskip, u32 *tailskip)
+gss_krb5_aes_decrypt(struct krb5_ctx *kctx, u32 offset, u32 len,
+		     struct xdr_buf *buf, u32 *headskip, u32 *tailskip)
 {
 	struct xdr_buf subbuf;
 	u32 ret = 0;
@@ -881,7 +881,7 @@ gss_krb5_aes_decrypt(struct krb5_ctx *kctx, u32 offset, struct xdr_buf *buf,
 
 	/* create a segment skipping the header and leaving out the checksum */
 	xdr_buf_subsegment(buf, &subbuf, offset + GSS_KRB5_TOK_HDR_LEN,
-				    (buf->len - offset - GSS_KRB5_TOK_HDR_LEN -
+				    (len - offset - GSS_KRB5_TOK_HDR_LEN -
 				     kctx->gk5e->cksumlength));
 
 	nblocks = (subbuf.len + blocksize - 1) / blocksize;
@@ -926,7 +926,7 @@ gss_krb5_aes_decrypt(struct krb5_ctx *kctx, u32 offset, struct xdr_buf *buf,
 		goto out_err;
 
 	/* Get the packet's hmac value */
-	ret = read_bytes_from_xdr_buf(buf, buf->len - kctx->gk5e->cksumlength,
+	ret = read_bytes_from_xdr_buf(buf, len - kctx->gk5e->cksumlength,
 				      pkt_hmac, kctx->gk5e->cksumlength);
 	if (ret)
 		goto out_err;
diff --git a/net/sunrpc/auth_gss/gss_krb5_wrap.c b/net/sunrpc/auth_gss/gss_krb5_wrap.c
index 6c1920eed771..cf0fd170ac18 100644
--- a/net/sunrpc/auth_gss/gss_krb5_wrap.c
+++ b/net/sunrpc/auth_gss/gss_krb5_wrap.c
@@ -261,7 +261,9 @@ gss_wrap_kerberos_v1(struct krb5_ctx *kctx, int offset,
 }
 
 static u32
-gss_unwrap_kerberos_v1(struct krb5_ctx *kctx, int offset, struct xdr_buf *buf)
+gss_unwrap_kerberos_v1(struct krb5_ctx *kctx, int offset, int len,
+		       struct xdr_buf *buf, unsigned int *slack,
+		       unsigned int *align)
 {
 	int			signalg;
 	int			sealalg;
@@ -279,12 +281,13 @@ gss_unwrap_kerberos_v1(struct krb5_ctx *kctx, int offset, struct xdr_buf *buf)
 	u32			conflen = kctx->gk5e->conflen;
 	int			crypt_offset;
 	u8			*cksumkey;
+	unsigned int		saved_len = buf->len;
 
 	dprintk("RPC:       gss_unwrap_kerberos\n");
 
 	ptr = (u8 *)buf->head[0].iov_base + offset;
 	if (g_verify_token_header(&kctx->mech_used, &bodysize, &ptr,
-					buf->len - offset))
+					len - offset))
 		return GSS_S_DEFECTIVE_TOKEN;
 
 	if ((ptr[0] != ((KG_TOK_WRAP_MSG >> 8) & 0xff)) ||
@@ -324,6 +327,7 @@ gss_unwrap_kerberos_v1(struct krb5_ctx *kctx, int offset, struct xdr_buf *buf)
 	    (!kctx->initiate && direction != 0))
 		return GSS_S_BAD_SIG;
 
+	buf->len = len;
 	if (kctx->enctype == ENCTYPE_ARCFOUR_HMAC) {
 		struct crypto_sync_skcipher *cipher;
 		int err;
@@ -376,11 +380,15 @@ gss_unwrap_kerberos_v1(struct krb5_ctx *kctx, int offset, struct xdr_buf *buf)
 	data_len = (buf->head[0].iov_base + buf->head[0].iov_len) - data_start;
 	memmove(orig_start, data_start, data_len);
 	buf->head[0].iov_len -= (data_start - orig_start);
-	buf->len -= (data_start - orig_start);
+	buf->len = len - (data_start - orig_start);
 
 	if (gss_krb5_remove_padding(buf, blocksize))
 		return GSS_S_DEFECTIVE_TOKEN;
 
+	/* slack must include room for krb5 padding */
+	*slack = XDR_QUADLEN(saved_len - buf->len);
+	/* The GSS blob always precedes the RPC message payload */
+	*align = *slack;
 	return GSS_S_COMPLETE;
 }
 
@@ -486,7 +494,9 @@ gss_wrap_kerberos_v2(struct krb5_ctx *kctx, u32 offset,
 }
 
 static u32
-gss_unwrap_kerberos_v2(struct krb5_ctx *kctx, int offset, struct xdr_buf *buf)
+gss_unwrap_kerberos_v2(struct krb5_ctx *kctx, int offset, int len,
+		       struct xdr_buf *buf, unsigned int *slack,
+		       unsigned int *align)
 {
 	time64_t	now;
 	u8		*ptr;
@@ -532,7 +542,7 @@ gss_unwrap_kerberos_v2(struct krb5_ctx *kctx, int offset, struct xdr_buf *buf)
 	if (rrc != 0)
 		rotate_left(offset + 16, buf, rrc);
 
-	err = (*kctx->gk5e->decrypt_v2)(kctx, offset, buf,
+	err = (*kctx->gk5e->decrypt_v2)(kctx, offset, len, buf,
 					&headskip, &tailskip);
 	if (err)
 		return GSS_S_FAILURE;
@@ -542,7 +552,7 @@ gss_unwrap_kerberos_v2(struct krb5_ctx *kctx, int offset, struct xdr_buf *buf)
 	 * it against the original
 	 */
 	err = read_bytes_from_xdr_buf(buf,
-				buf->len - GSS_KRB5_TOK_HDR_LEN - tailskip,
+				len - GSS_KRB5_TOK_HDR_LEN - tailskip,
 				decrypted_hdr, GSS_KRB5_TOK_HDR_LEN);
 	if (err) {
 		dprintk("%s: error %u getting decrypted_hdr\n", __func__, err);
@@ -568,18 +578,19 @@ gss_unwrap_kerberos_v2(struct krb5_ctx *kctx, int offset, struct xdr_buf *buf)
 	 * Note that buf->head[0].iov_len may indicate the available
 	 * head buffer space rather than that actually occupied.
 	 */
-	movelen = min_t(unsigned int, buf->head[0].iov_len, buf->len);
+	movelen = min_t(unsigned int, buf->head[0].iov_len, len);
 	movelen -= offset + GSS_KRB5_TOK_HDR_LEN + headskip;
-	if (offset + GSS_KRB5_TOK_HDR_LEN + headskip + movelen >
-	    buf->head[0].iov_len)
-		return GSS_S_FAILURE;
+	BUG_ON(offset + GSS_KRB5_TOK_HDR_LEN + headskip + movelen >
+							buf->head[0].iov_len);
 	memmove(ptr, ptr + GSS_KRB5_TOK_HDR_LEN + headskip, movelen);
 	buf->head[0].iov_len -= GSS_KRB5_TOK_HDR_LEN + headskip;
-	buf->len -= GSS_KRB5_TOK_HDR_LEN + headskip;
+	buf->len = len - GSS_KRB5_TOK_HDR_LEN + headskip;
 
 	/* Trim off the trailing "extra count" and checksum blob */
-	buf->len -= ec + GSS_KRB5_TOK_HDR_LEN + tailskip;
+	xdr_buf_trim(buf, ec + GSS_KRB5_TOK_HDR_LEN + tailskip);
 
+	*align = XDR_QUADLEN(GSS_KRB5_TOK_HDR_LEN + headskip);
+	*slack = *align + XDR_QUADLEN(ec + GSS_KRB5_TOK_HDR_LEN + tailskip);
 	return GSS_S_COMPLETE;
 }
 
@@ -603,7 +614,8 @@ gss_wrap_kerberos(struct gss_ctx *gctx, int offset,
 }
 
 u32
-gss_unwrap_kerberos(struct gss_ctx *gctx, int offset, struct xdr_buf *buf)
+gss_unwrap_kerberos(struct gss_ctx *gctx, int offset,
+		    int len, struct xdr_buf *buf)
 {
 	struct krb5_ctx	*kctx = gctx->internal_ctx_id;
 
@@ -613,9 +625,11 @@ gss_unwrap_kerberos(struct gss_ctx *gctx, int offset, struct xdr_buf *buf)
 	case ENCTYPE_DES_CBC_RAW:
 	case ENCTYPE_DES3_CBC_RAW:
 	case ENCTYPE_ARCFOUR_HMAC:
-		return gss_unwrap_kerberos_v1(kctx, offset, buf);
+		return gss_unwrap_kerberos_v1(kctx, offset, len, buf,
+					      &gctx->slack, &gctx->align);
 	case ENCTYPE_AES128_CTS_HMAC_SHA1_96:
 	case ENCTYPE_AES256_CTS_HMAC_SHA1_96:
-		return gss_unwrap_kerberos_v2(kctx, offset, buf);
+		return gss_unwrap_kerberos_v2(kctx, offset, len, buf,
+					      &gctx->slack, &gctx->align);
 	}
 }
diff --git a/net/sunrpc/auth_gss/gss_mech_switch.c b/net/sunrpc/auth_gss/gss_mech_switch.c
index db550bfc2642..69316ab1b9fa 100644
--- a/net/sunrpc/auth_gss/gss_mech_switch.c
+++ b/net/sunrpc/auth_gss/gss_mech_switch.c
@@ -411,10 +411,11 @@ gss_wrap(struct gss_ctx	*ctx_id,
 u32
 gss_unwrap(struct gss_ctx	*ctx_id,
 	   int			offset,
+	   int			len,
 	   struct xdr_buf	*buf)
 {
 	return ctx_id->mech_type->gm_ops
-		->gss_unwrap(ctx_id, offset, buf);
+		->gss_unwrap(ctx_id, offset, len, buf);
 }
 
 
diff --git a/net/sunrpc/auth_gss/svcauth_gss.c b/net/sunrpc/auth_gss/svcauth_gss.c
index 54ae5be62f6a..50d93c49ef1a 100644
--- a/net/sunrpc/auth_gss/svcauth_gss.c
+++ b/net/sunrpc/auth_gss/svcauth_gss.c
@@ -906,7 +906,7 @@ unwrap_integ_data(struct svc_rqst *rqstp, struct xdr_buf *buf, u32 seq, struct g
 	if (svc_getnl(&buf->head[0]) != seq)
 		goto out;
 	/* trim off the mic and padding at the end before returning */
-	buf->len -= 4 + round_up_to_quad(mic.len);
+	xdr_buf_trim(buf, round_up_to_quad(mic.len) + 4);
 	stat = 0;
 out:
 	kfree(mic.data);
@@ -934,7 +934,7 @@ static int
 unwrap_priv_data(struct svc_rqst *rqstp, struct xdr_buf *buf, u32 seq, struct gss_ctx *ctx)
 {
 	u32 priv_len, maj_stat;
-	int pad, saved_len, remaining_len, offset;
+	int pad, remaining_len, offset;
 
 	clear_bit(RQ_SPLICE_OK, &rqstp->rq_flags);
 
@@ -954,12 +954,8 @@ unwrap_priv_data(struct svc_rqst *rqstp, struct xdr_buf *buf, u32 seq, struct gs
 	buf->len -= pad;
 	fix_priv_head(buf, pad);
 
-	/* Maybe it would be better to give gss_unwrap a length parameter: */
-	saved_len = buf->len;
-	buf->len = priv_len;
-	maj_stat = gss_unwrap(ctx, 0, buf);
+	maj_stat = gss_unwrap(ctx, 0, priv_len, buf);
 	pad = priv_len - buf->len;
-	buf->len = saved_len;
 	buf->len -= pad;
 	/* The upper layers assume the buffer is aligned on 4-byte boundaries.
 	 * In the krb5p case, at least, the data ends up offset, so we need to
diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c
index 15b58c5144f9..6f7d82fb1eb0 100644
--- a/net/sunrpc/xdr.c
+++ b/net/sunrpc/xdr.c
@@ -1150,6 +1150,47 @@ xdr_buf_subsegment(struct xdr_buf *buf, struct xdr_buf *subbuf,
 }
 EXPORT_SYMBOL_GPL(xdr_buf_subsegment);
 
+/**
+ * xdr_buf_trim - lop at most "len" bytes off the end of "buf"
+ * @buf: buf to be trimmed
+ * @len: number of bytes to reduce "buf" by
+ *
+ * Trim an xdr_buf by the given number of bytes by fixing up the lengths. Note
+ * that it's possible that we'll trim less than that amount if the xdr_buf is
+ * too small, or if (for instance) it's all in the head and the parser has
+ * already read too far into it.
+ */
+void xdr_buf_trim(struct xdr_buf *buf, unsigned int len)
+{
+	size_t cur;
+	unsigned int trim = len;
+
+	if (buf->tail[0].iov_len) {
+		cur = min_t(size_t, buf->tail[0].iov_len, trim);
+		buf->tail[0].iov_len -= cur;
+		trim -= cur;
+		if (!trim)
+			goto fix_len;
+	}
+
+	if (buf->page_len) {
+		cur = min_t(unsigned int, buf->page_len, trim);
+		buf->page_len -= cur;
+		trim -= cur;
+		if (!trim)
+			goto fix_len;
+	}
+
+	if (buf->head[0].iov_len) {
+		cur = min_t(size_t, buf->head[0].iov_len, trim);
+		buf->head[0].iov_len -= cur;
+		trim -= cur;
+	}
+fix_len:
+	buf->len -= (len - trim);
+}
+EXPORT_SYMBOL_GPL(xdr_buf_trim);
+
 static void __read_bytes_from_xdr_buf(struct xdr_buf *subbuf, void *obj, unsigned int len)
 {
 	unsigned int this_len;
diff --git a/net/tipc/socket.c b/net/tipc/socket.c
index 87466607097f..e370ad0edd76 100644
--- a/net/tipc/socket.c
+++ b/net/tipc/socket.c
@@ -1739,22 +1739,21 @@ static int tipc_sk_anc_data_recv(struct msghdr *m, struct sk_buff *skb,
 	return 0;
 }
 
-static void tipc_sk_send_ack(struct tipc_sock *tsk)
+static struct sk_buff *tipc_sk_build_ack(struct tipc_sock *tsk)
 {
 	struct sock *sk = &tsk->sk;
-	struct net *net = sock_net(sk);
 	struct sk_buff *skb = NULL;
 	struct tipc_msg *msg;
 	u32 peer_port = tsk_peer_port(tsk);
 	u32 dnode = tsk_peer_node(tsk);
 
 	if (!tipc_sk_connected(sk))
-		return;
+		return NULL;
 	skb = tipc_msg_create(CONN_MANAGER, CONN_ACK, INT_H_SIZE, 0,
 			      dnode, tsk_own_node(tsk), peer_port,
 			      tsk->portid, TIPC_OK);
 	if (!skb)
-		return;
+		return NULL;
 	msg = buf_msg(skb);
 	msg_set_conn_ack(msg, tsk->rcv_unacked);
 	tsk->rcv_unacked = 0;
@@ -1764,7 +1763,19 @@ static void tipc_sk_send_ack(struct tipc_sock *tsk)
 		tsk->rcv_win = tsk_adv_blocks(tsk->sk.sk_rcvbuf);
 		msg_set_adv_win(msg, tsk->rcv_win);
 	}
-	tipc_node_xmit_skb(net, skb, dnode, msg_link_selector(msg));
+	return skb;
+}
+
+static void tipc_sk_send_ack(struct tipc_sock *tsk)
+{
+	struct sk_buff *skb;
+
+	skb = tipc_sk_build_ack(tsk);
+	if (!skb)
+		return;
+
+	tipc_node_xmit_skb(sock_net(&tsk->sk), skb, tsk_peer_node(tsk),
+			   msg_link_selector(buf_msg(skb)));
 }
 
 static int tipc_wait_for_rcvmsg(struct socket *sock, long *timeop)
@@ -1938,7 +1949,6 @@ static int tipc_recvstream(struct socket *sock, struct msghdr *m,
 	bool peek = flags & MSG_PEEK;
 	int offset, required, copy, copied = 0;
 	int hlen, dlen, err, rc;
-	bool ack = false;
 	long timeout;
 
 	/* Catch invalid receive attempts */
@@ -1983,7 +1993,6 @@ static int tipc_recvstream(struct socket *sock, struct msghdr *m,
 
 		/* Copy data if msg ok, otherwise return error/partial data */
 		if (likely(!err)) {
-			ack = msg_ack_required(hdr);
 			offset = skb_cb->bytes_read;
 			copy = min_t(int, dlen - offset, buflen - copied);
 			rc = skb_copy_datagram_msg(skb, hlen + offset, m, copy);
@@ -2011,7 +2020,7 @@ static int tipc_recvstream(struct socket *sock, struct msghdr *m,
 
 		/* Send connection flow control advertisement when applicable */
 		tsk->rcv_unacked += tsk_inc(tsk, hlen + dlen);
-		if (ack || tsk->rcv_unacked >= tsk->rcv_win / TIPC_ACK_RATE)
+		if (tsk->rcv_unacked >= tsk->rcv_win / TIPC_ACK_RATE)
 			tipc_sk_send_ack(tsk);
 
 		/* Exit if all requested data or FIN/error received */
@@ -2105,9 +2114,11 @@ static void tipc_sk_proto_rcv(struct sock *sk,
  * tipc_sk_filter_connect - check incoming message for a connection-based socket
  * @tsk: TIPC socket
  * @skb: pointer to message buffer.
+ * @xmitq: for Nagle ACK if any
  * Returns true if message should be added to receive queue, false otherwise
  */
-static bool tipc_sk_filter_connect(struct tipc_sock *tsk, struct sk_buff *skb)
+static bool tipc_sk_filter_connect(struct tipc_sock *tsk, struct sk_buff *skb,
+				   struct sk_buff_head *xmitq)
 {
 	struct sock *sk = &tsk->sk;
 	struct net *net = sock_net(sk);
@@ -2171,8 +2182,17 @@ static bool tipc_sk_filter_connect(struct tipc_sock *tsk, struct sk_buff *skb)
 		if (!skb_queue_empty(&sk->sk_write_queue))
 			tipc_sk_push_backlog(tsk);
 		/* Accept only connection-based messages sent by peer */
-		if (likely(con_msg && !err && pport == oport && pnode == onode))
+		if (likely(con_msg && !err && pport == oport &&
+			   pnode == onode)) {
+			if (msg_ack_required(hdr)) {
+				struct sk_buff *skb;
+
+				skb = tipc_sk_build_ack(tsk);
+				if (skb)
+					__skb_queue_tail(xmitq, skb);
+			}
 			return true;
+		}
 		if (!tsk_peer_msg(tsk, hdr))
 			return false;
 		if (!err)
@@ -2267,7 +2287,7 @@ static void tipc_sk_filter_rcv(struct sock *sk, struct sk_buff *skb,
 	while ((skb = __skb_dequeue(&inputq))) {
 		hdr = buf_msg(skb);
 		limit = rcvbuf_limit(sk, skb);
-		if ((sk_conn && !tipc_sk_filter_connect(tsk, skb)) ||
+		if ((sk_conn && !tipc_sk_filter_connect(tsk, skb, xmitq)) ||
 		    (!sk_conn && msg_connected(hdr)) ||
 		    (!grp && msg_in_group(hdr)))
 			err = TIPC_ERR_NO_PORT;
diff --git a/net/tipc/subscr.h b/net/tipc/subscr.h
index aa015c233898..6ebbec1bedd1 100644
--- a/net/tipc/subscr.h
+++ b/net/tipc/subscr.h
@@ -96,6 +96,16 @@ void tipc_sub_get(struct tipc_subscription *subscription);
 		(swap_ ? swab32(val__) : val__);			\
 	})
 
+/* tipc_sub_write - write val_ to field_ of struct sub_ in user endian format
+ */
+#define tipc_sub_write(sub_, field_, val_)				\
+	({								\
+		struct tipc_subscr *sub__ = sub_;			\
+		u32 val__ = val_;					\
+		int swap_ = !((sub__)->filter & TIPC_FILTER_MASK);	\
+		(sub__)->field_ = swap_ ? swab32(val__) : val__;	\
+	})
+
 /* tipc_evt_write - write val_ to field_ of struct evt_ in user endian format
  */
 #define tipc_evt_write(evt_, field_, val_)				\
diff --git a/net/tipc/topsrv.c b/net/tipc/topsrv.c
index 73dbed0c4b6b..446af7bbd13e 100644
--- a/net/tipc/topsrv.c
+++ b/net/tipc/topsrv.c
@@ -237,8 +237,8 @@ static void tipc_conn_delete_sub(struct tipc_conn *con, struct tipc_subscr *s)
 		if (!s || !memcmp(s, &sub->evt.s, sizeof(*s))) {
 			tipc_sub_unsubscribe(sub);
 			atomic_dec(&tn->subscription_count);
-		} else if (s) {
-			break;
+			if (s)
+				break;
 		}
 	}
 	spin_unlock_bh(&con->sub_lock);
@@ -362,9 +362,10 @@ static int tipc_conn_rcv_sub(struct tipc_topsrv *srv,
 {
 	struct tipc_net *tn = tipc_net(srv->net);
 	struct tipc_subscription *sub;
+	u32 s_filter = tipc_sub_read(s, filter);
 
-	if (tipc_sub_read(s, filter) & TIPC_SUB_CANCEL) {
-		s->filter &= __constant_ntohl(~TIPC_SUB_CANCEL);
+	if (s_filter & TIPC_SUB_CANCEL) {
+		tipc_sub_write(s, filter, s_filter & ~TIPC_SUB_CANCEL);
 		tipc_conn_delete_sub(con, s);
 		return 0;
 	}
@@ -400,7 +401,9 @@ static int tipc_conn_rcv_from_sock(struct tipc_conn *con)
 		return -EWOULDBLOCK;
 	if (ret == sizeof(s)) {
 		read_lock_bh(&sk->sk_callback_lock);
-		ret = tipc_conn_rcv_sub(srv, con, &s);
+		/* RACE: the connection can be closed in the meantime */
+		if (likely(connected(con)))
+			ret = tipc_conn_rcv_sub(srv, con, &s);
 		read_unlock_bh(&sk->sk_callback_lock);
 		if (!ret)
 			return 0;