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

Pull networking fixes from David Miller:

 1) Fix several scatter gather list issues in kTLS code, from Jakub
    Kicinski.

 2) macb driver device remove has to kill the hresp_err_tasklet. From
    Chuhong Yuan.

 3) Several memory leak and reference count bug fixes in tipc, from Tung
    Nguyen.

 4) Fix mlx5 build error w/o ipv6, from Yue Haibing.

 5) Fix jumbo frame and other regressions in r8169, from Heiner
    Kallweit.

 6) Undo some BUG_ON()'s and replace them with WARN_ON_ONCE and proper
    error propagation/handling. From Paolo Abeni.

* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (24 commits)
  openvswitch: remove another BUG_ON()
  openvswitch: drop unneeded BUG_ON() in ovs_flow_cmd_build_info()
  net: phy: realtek: fix using paged operations with RTL8105e / RTL8208
  r8169: fix resume on cable plug-in
  r8169: fix jumbo configuration for RTL8168evl
  net: emulex: benet: indent a Kconfig depends continuation line
  selftests: forwarding: fix race between packet receive and tc check
  net: sched: fix `tc -s class show` no bstats on class with nolock subqueues
  net: ethernet: ti: ale: ensure vlan/mdb deleted when no members
  net/mlx5e: Fix build error without IPV6
  selftests: pmtu: use -oneline for ip route list cache
  tipc: fix duplicate SYN messages under link congestion
  tipc: fix wrong timeout input for tipc_wait_for_cond()
  tipc: fix wrong socket reference counter after tipc_sk_timeout() returns
  tipc: fix potential memory leak in __tipc_sendmsg()
  net: macb: add missed tasklet_kill
  selftests: bpf: correct perror strings
  selftests: bpf: test_sockmap: handle file creation failures gracefully
  net/tls: use sg_next() to walk sg entries
  net/tls: remove the dead inplace_crypto code
  ...

drivers/net/ethernet/cadence/macb_main.c

@@ -4422,6 +4422,7 @@ static int macb_remove(struct platform_device *pdev)
 		mdiobus_free(bp->mii_bus);
 
 		unregister_netdev(dev);
+		tasklet_kill(&bp->hresp_err_tasklet);
 		pm_runtime_disable(&pdev->dev);
 		pm_runtime_dont_use_autosuspend(&pdev->dev);
 		if (!pm_runtime_suspended(&pdev->dev)) {

drivers/net/ethernet/emulex/benet/Kconfig

@@ -49,4 +49,4 @@ config BE2NET_SKYHAWK
 
 comment "WARNING: be2net is useless without any enabled chip"
 	depends on BE2NET_BE2=n && BE2NET_BE3=n && BE2NET_LANCER=n && \
-	BE2NET_SKYHAWK=n && BE2NET
+		BE2NET_SKYHAWK=n && BE2NET

drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c

@@ -130,42 +130,6 @@ static const char *mlx5e_netdev_kind(struct net_device *dev)
 		return "unknown";
 }
 
-static int mlx5e_route_lookup_ipv6(struct mlx5e_priv *priv,
-				   struct net_device *mirred_dev,
-				   struct net_device **out_dev,
-				   struct net_device **route_dev,
-				   struct flowi6 *fl6,
-				   struct neighbour **out_n,
-				   u8 *out_ttl)
-{
-	struct dst_entry *dst;
-	struct neighbour *n;
-
-	int ret;
-
-	ret = ipv6_stub->ipv6_dst_lookup(dev_net(mirred_dev), NULL, &dst,
-					 fl6);
-	if (ret < 0)
-		return ret;
-
-	if (!(*out_ttl))
-		*out_ttl = ip6_dst_hoplimit(dst);
-
-	ret = get_route_and_out_devs(priv, dst->dev, route_dev, out_dev);
-	if (ret < 0) {
-		dst_release(dst);
-		return ret;
-	}
-
-	n = dst_neigh_lookup(dst, &fl6->daddr);
-	dst_release(dst);
-	if (!n)
-		return -ENOMEM;
-
-	*out_n = n;
-	return 0;
-}
-
 static int mlx5e_gen_ip_tunnel_header(char buf[], __u8 *ip_proto,
 				      struct mlx5e_encap_entry *e)
 {
@@ -319,6 +283,43 @@ int mlx5e_tc_tun_create_header_ipv4(struct mlx5e_priv *priv,
 	return err;
 }
 
+#if IS_ENABLED(CONFIG_INET) && IS_ENABLED(CONFIG_IPV6)
+static int mlx5e_route_lookup_ipv6(struct mlx5e_priv *priv,
+				   struct net_device *mirred_dev,
+				   struct net_device **out_dev,
+				   struct net_device **route_dev,
+				   struct flowi6 *fl6,
+				   struct neighbour **out_n,
+				   u8 *out_ttl)
+{
+	struct dst_entry *dst;
+	struct neighbour *n;
+
+	int ret;
+
+	ret = ipv6_stub->ipv6_dst_lookup(dev_net(mirred_dev), NULL, &dst,
+					 fl6);
+	if (ret < 0)
+		return ret;
+
+	if (!(*out_ttl))
+		*out_ttl = ip6_dst_hoplimit(dst);
+
+	ret = get_route_and_out_devs(priv, dst->dev, route_dev, out_dev);
+	if (ret < 0) {
+		dst_release(dst);
+		return ret;
+	}
+
+	n = dst_neigh_lookup(dst, &fl6->daddr);
+	dst_release(dst);
+	if (!n)
+		return -ENOMEM;
+
+	*out_n = n;
+	return 0;
+}
+
 int mlx5e_tc_tun_create_header_ipv6(struct mlx5e_priv *priv,
 				    struct net_device *mirred_dev,
 				    struct mlx5e_encap_entry *e)
@@ -436,6 +437,7 @@ int mlx5e_tc_tun_create_header_ipv6(struct mlx5e_priv *priv,
 	neigh_release(n);
 	return err;
 }
+#endif
 
 bool mlx5e_tc_tun_device_to_offload(struct mlx5e_priv *priv,
 				    struct net_device *netdev)

drivers/net/ethernet/realtek/r8169_main.c

@@ -1542,6 +1542,7 @@ static void __rtl8169_set_wol(struct rtl8169_private *tp, u32 wolopts)
 	rtl_lock_config_regs(tp);
 
 	device_set_wakeup_enable(tp_to_dev(tp), wolopts);
+	tp->dev->wol_enabled = wolopts ? 1 : 0;
 }
 
 static int rtl8169_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
@@ -3872,7 +3873,7 @@ static void rtl_hw_jumbo_enable(struct rtl8169_private *tp)
 	case RTL_GIGA_MAC_VER_27 ... RTL_GIGA_MAC_VER_28:
 		r8168dp_hw_jumbo_enable(tp);
 		break;
-	case RTL_GIGA_MAC_VER_31 ... RTL_GIGA_MAC_VER_34:
+	case RTL_GIGA_MAC_VER_31 ... RTL_GIGA_MAC_VER_33:
 		r8168e_hw_jumbo_enable(tp);
 		break;
 	default:

drivers/net/ethernet/ti/cpsw_ale.c

@@ -384,7 +384,7 @@ int cpsw_ale_del_mcast(struct cpsw_ale *ale, const u8 *addr, int port_mask,
 		       int flags, u16 vid)
 {
 	u32 ale_entry[ALE_ENTRY_WORDS] = {0, 0, 0};
-	int mcast_members;
+	int mcast_members = 0;
 	int idx;
 
 	idx = cpsw_ale_match_addr(ale, addr, (flags & ALE_VLAN) ? vid : 0);
@@ -397,11 +397,13 @@ int cpsw_ale_del_mcast(struct cpsw_ale *ale, const u8 *addr, int port_mask,
 		mcast_members = cpsw_ale_get_port_mask(ale_entry,
 						       ale->port_mask_bits);
 		mcast_members &= ~port_mask;
+	}
+
+	if (mcast_members)
 		cpsw_ale_set_port_mask(ale_entry, mcast_members,
 				       ale->port_mask_bits);
-	} else {
+	else
 		cpsw_ale_set_entry_type(ale_entry, ALE_TYPE_FREE);
-	}
 
 	cpsw_ale_write(ale, idx, ale_entry);
 	return 0;
@@ -478,6 +480,10 @@ static void cpsw_ale_del_vlan_modify(struct cpsw_ale *ale, u32 *ale_entry,
 	members = cpsw_ale_get_vlan_member_list(ale_entry,
 						ale->vlan_field_bits);
 	members &= ~port_mask;
+	if (!members) {
+		cpsw_ale_set_entry_type(ale_entry, ALE_TYPE_FREE);
+		return;
+	}
 
 	untag = cpsw_ale_get_vlan_untag_force(ale_entry,
 					      ale->vlan_field_bits);

drivers/net/phy/realtek.c

@@ -439,6 +439,15 @@ static struct phy_driver realtek_drvs[] = {
 		.resume		= genphy_resume,
 		.read_page	= rtl821x_read_page,
 		.write_page	= rtl821x_write_page,
+	}, {
+		PHY_ID_MATCH_MODEL(0x001cc880),
+		.name		= "RTL8208 Fast Ethernet",
+		.read_mmd	= genphy_read_mmd_unsupported,
+		.write_mmd	= genphy_write_mmd_unsupported,
+		.suspend	= genphy_suspend,
+		.resume		= genphy_resume,
+		.read_page	= rtl821x_read_page,
+		.write_page	= rtl821x_write_page,
 	}, {
 		PHY_ID_MATCH_EXACT(0x001cc910),
 		.name		= "RTL8211 Gigabit Ethernet",

include/linux/skmsg.h

@@ -14,6 +14,7 @@
 #include <net/strparser.h>
 
 #define MAX_MSG_FRAGS			MAX_SKB_FRAGS
+#define NR_MSG_FRAG_IDS			(MAX_MSG_FRAGS + 1)
 
 enum __sk_action {
 	__SK_DROP = 0,
@@ -29,13 +30,15 @@ struct sk_msg_sg {
 	u32				size;
 	u32				copybreak;
 	unsigned long			copy;
-	/* The extra element is used for chaining the front and sections when
-	 * the list becomes partitioned (e.g. end < start). The crypto APIs
-	 * require the chaining.
+	/* The extra two elements:
+	 * 1) used for chaining the front and sections when the list becomes
+	 *    partitioned (e.g. end < start). The crypto APIs require the
+	 *    chaining;
+	 * 2) to chain tailer SG entries after the message.
 	 */
-	struct scatterlist		data[MAX_MSG_FRAGS + 1];
+	struct scatterlist		data[MAX_MSG_FRAGS + 2];
 };
-static_assert(BITS_PER_LONG >= MAX_MSG_FRAGS);
+static_assert(BITS_PER_LONG >= NR_MSG_FRAG_IDS);
 
 /* UAPI in filter.c depends on struct sk_msg_sg being first element. */
 struct sk_msg {
@@ -142,13 +145,13 @@ static inline void sk_msg_apply_bytes(struct sk_psock *psock, u32 bytes)
 
 static inline u32 sk_msg_iter_dist(u32 start, u32 end)
 {
-	return end >= start ? end - start : end + (MAX_MSG_FRAGS - start);
+	return end >= start ? end - start : end + (NR_MSG_FRAG_IDS - start);
 }
 
 #define sk_msg_iter_var_prev(var)			\
 	do {						\
 		if (var == 0)				\
-			var = MAX_MSG_FRAGS - 1;	\
+			var = NR_MSG_FRAG_IDS - 1;	\
 		else					\
 			var--;				\
 	} while (0)
@@ -156,7 +159,7 @@ static inline u32 sk_msg_iter_dist(u32 start, u32 end)
 #define sk_msg_iter_var_next(var)			\
 	do {						\
 		var++;					\
-		if (var == MAX_MSG_FRAGS)		\
+		if (var == NR_MSG_FRAG_IDS)		\
 			var = 0;			\
 	} while (0)
 
@@ -173,9 +176,9 @@ static inline void sk_msg_clear_meta(struct sk_msg *msg)
 
 static inline void sk_msg_init(struct sk_msg *msg)
 {
-	BUILD_BUG_ON(ARRAY_SIZE(msg->sg.data) - 1 != MAX_MSG_FRAGS);
+	BUILD_BUG_ON(ARRAY_SIZE(msg->sg.data) - 1 != NR_MSG_FRAG_IDS);
 	memset(msg, 0, sizeof(*msg));
-	sg_init_marker(msg->sg.data, MAX_MSG_FRAGS);
+	sg_init_marker(msg->sg.data, NR_MSG_FRAG_IDS);
 }
 
 static inline void sk_msg_xfer(struct sk_msg *dst, struct sk_msg *src,
@@ -196,14 +199,11 @@ static inline void sk_msg_xfer_full(struct sk_msg *dst, struct sk_msg *src)
 
 static inline bool sk_msg_full(const struct sk_msg *msg)
 {
-	return (msg->sg.end == msg->sg.start) && msg->sg.size;
+	return sk_msg_iter_dist(msg->sg.start, msg->sg.end) == MAX_MSG_FRAGS;
 }
 
 static inline u32 sk_msg_elem_used(const struct sk_msg *msg)
 {
-	if (sk_msg_full(msg))
-		return MAX_MSG_FRAGS;
-
 	return sk_msg_iter_dist(msg->sg.start, msg->sg.end);
 }
 

include/net/tls.h

@@ -100,7 +100,6 @@ struct tls_rec {
 	struct list_head list;
 	int tx_ready;
 	int tx_flags;
-	int inplace_crypto;
 
 	struct sk_msg msg_plaintext;
 	struct sk_msg msg_encrypted;
@@ -377,7 +376,7 @@ int tls_push_sg(struct sock *sk, struct tls_context *ctx,
 		int flags);
 int tls_push_partial_record(struct sock *sk, struct tls_context *ctx,
 			    int flags);
-bool tls_free_partial_record(struct sock *sk, struct tls_context *ctx);
+void tls_free_partial_record(struct sock *sk, struct tls_context *ctx);
 
 static inline struct tls_msg *tls_msg(struct sk_buff *skb)
 {

net/core/filter.c

@@ -2299,7 +2299,7 @@ BPF_CALL_4(bpf_msg_pull_data, struct sk_msg *, msg, u32, start,
 	WARN_ON_ONCE(last_sge == first_sge);
 	shift = last_sge > first_sge ?
 		last_sge - first_sge - 1 :
-		MAX_SKB_FRAGS - first_sge + last_sge - 1;
+		NR_MSG_FRAG_IDS - first_sge + last_sge - 1;
 	if (!shift)
 		goto out;
 
@@ -2308,8 +2308,8 @@ BPF_CALL_4(bpf_msg_pull_data, struct sk_msg *, msg, u32, start,
 	do {
 		u32 move_from;
 
-		if (i + shift >= MAX_MSG_FRAGS)
-			move_from = i + shift - MAX_MSG_FRAGS;
+		if (i + shift >= NR_MSG_FRAG_IDS)
+			move_from = i + shift - NR_MSG_FRAG_IDS;
 		else
 			move_from = i + shift;
 		if (move_from == msg->sg.end)
@@ -2323,7 +2323,7 @@ BPF_CALL_4(bpf_msg_pull_data, struct sk_msg *, msg, u32, start,
 	} while (1);
 
 	msg->sg.end = msg->sg.end - shift > msg->sg.end ?
-		      msg->sg.end - shift + MAX_MSG_FRAGS :
+		      msg->sg.end - shift + NR_MSG_FRAG_IDS :
 		      msg->sg.end - shift;
 out:
 	msg->data = sg_virt(&msg->sg.data[first_sge]) + start - offset;

net/core/skmsg.c

@@ -421,7 +421,7 @@ static int sk_psock_skb_ingress(struct sk_psock *psock, struct sk_buff *skb)
 	copied = skb->len;
 	msg->sg.start = 0;
 	msg->sg.size = copied;
-	msg->sg.end = num_sge == MAX_MSG_FRAGS ? 0 : num_sge;
+	msg->sg.end = num_sge;
 	msg->skb = skb;
 
 	sk_psock_queue_msg(psock, msg);

net/ipv4/tcp_bpf.c

@@ -301,7 +301,7 @@ EXPORT_SYMBOL_GPL(tcp_bpf_sendmsg_redir);
 static int tcp_bpf_send_verdict(struct sock *sk, struct sk_psock *psock,
 				struct sk_msg *msg, int *copied, int flags)
 {
-	bool cork = false, enospc = msg->sg.start == msg->sg.end;
+	bool cork = false, enospc = sk_msg_full(msg);
 	struct sock *sk_redir;
 	u32 tosend, delta = 0;
 	int ret;

net/openvswitch/datapath.c

@@ -905,7 +905,10 @@ static struct sk_buff *ovs_flow_cmd_build_info(const struct sw_flow *flow,
 	retval = ovs_flow_cmd_fill_info(flow, dp_ifindex, skb,
 					info->snd_portid, info->snd_seq, 0,
 					cmd, ufid_flags);
-	BUG_ON(retval < 0);
+	if (WARN_ON_ONCE(retval < 0)) {
+		kfree_skb(skb);
+		skb = ERR_PTR(retval);
+	}
 	return skb;
 }
 
@@ -1369,7 +1372,10 @@ static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info)
 						     OVS_FLOW_CMD_DEL,
 						     ufid_flags);
 			rcu_read_unlock();
-			BUG_ON(err < 0);
+			if (WARN_ON_ONCE(err < 0)) {
+				kfree_skb(reply);
+				goto out_free;
+			}
 
 			ovs_notify(&dp_flow_genl_family, reply, info);
 		} else {
@@ -1377,6 +1383,7 @@ static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info)
 		}
 	}
 
+out_free:
 	ovs_flow_free(flow, true);
 	return 0;
 unlock:

net/sched/sch_mq.c

@@ -245,7 +245,8 @@ static int mq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
 	struct netdev_queue *dev_queue = mq_queue_get(sch, cl);
 
 	sch = dev_queue->qdisc_sleeping;
-	if (gnet_stats_copy_basic(&sch->running, d, NULL, &sch->bstats) < 0 ||
+	if (gnet_stats_copy_basic(&sch->running, d, sch->cpu_bstats,
+				  &sch->bstats) < 0 ||
 	    qdisc_qstats_copy(d, sch) < 0)
 		return -1;
 	return 0;

net/sched/sch_mqprio.c

@@ -557,8 +557,8 @@ static int mqprio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
 		struct netdev_queue *dev_queue = mqprio_queue_get(sch, cl);
 
 		sch = dev_queue->qdisc_sleeping;
-		if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
-					  d, NULL, &sch->bstats) < 0 ||
+		if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), d,
+					  sch->cpu_bstats, &sch->bstats) < 0 ||
 		    qdisc_qstats_copy(d, sch) < 0)
 			return -1;
 	}

net/sched/sch_multiq.c

@@ -339,7 +339,7 @@ static int multiq_dump_class_stats(struct Qdisc *sch, unsigned long cl,
 
 	cl_q = q->queues[cl - 1];
 	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
-				  d, NULL, &cl_q->bstats) < 0 ||
+				  d, cl_q->cpu_bstats, &cl_q->bstats) < 0 ||
 	    qdisc_qstats_copy(d, cl_q) < 0)
 		return -1;
 

net/sched/sch_prio.c

@@ -356,7 +356,7 @@ static int prio_dump_class_stats(struct Qdisc *sch, unsigned long cl,
 
 	cl_q = q->queues[cl - 1];
 	if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch),
-				  d, NULL, &cl_q->bstats) < 0 ||
+				  d, cl_q->cpu_bstats, &cl_q->bstats) < 0 ||
 	    qdisc_qstats_copy(d, cl_q) < 0)
 		return -1;
 

net/tipc/socket.c

@@ -532,7 +532,7 @@ static void __tipc_shutdown(struct socket *sock, int error)
 	struct sock *sk = sock->sk;
 	struct tipc_sock *tsk = tipc_sk(sk);
 	struct net *net = sock_net(sk);
-	long timeout = CONN_TIMEOUT_DEFAULT;
+	long timeout = msecs_to_jiffies(CONN_TIMEOUT_DEFAULT);
 	u32 dnode = tsk_peer_node(tsk);
 	struct sk_buff *skb;
 
@@ -540,12 +540,10 @@ static void __tipc_shutdown(struct socket *sock, int error)
 	tipc_wait_for_cond(sock, &timeout, (!tsk->cong_link_cnt &&
 					    !tsk_conn_cong(tsk)));
 
-	/* Push out unsent messages or remove if pending SYN */
-	skb = skb_peek(&sk->sk_write_queue);
-	if (skb && !msg_is_syn(buf_msg(skb)))
-		tipc_sk_push_backlog(tsk);
-	else
-		__skb_queue_purge(&sk->sk_write_queue);
+	/* Push out delayed messages if in Nagle mode */
+	tipc_sk_push_backlog(tsk);
+	/* Remove pending SYN */
+	__skb_queue_purge(&sk->sk_write_queue);
 
 	/* Reject all unreceived messages, except on an active connection
 	 * (which disconnects locally & sends a 'FIN+' to peer).
@@ -1248,9 +1246,14 @@ static void tipc_sk_push_backlog(struct tipc_sock *tsk)
 	struct sk_buff_head *txq = &tsk->sk.sk_write_queue;
 	struct net *net = sock_net(&tsk->sk);
 	u32 dnode = tsk_peer_node(tsk);
+	struct sk_buff *skb = skb_peek(txq);
 	int rc;
 
-	if (skb_queue_empty(txq) || tsk->cong_link_cnt)
+	if (!skb || tsk->cong_link_cnt)
+		return;
+
+	/* Do not send SYN again after congestion */
+	if (msg_is_syn(buf_msg(skb)))
 		return;
 
 	tsk->snt_unacked += tsk->snd_backlog;
@@ -1447,8 +1450,10 @@ static int __tipc_sendmsg(struct socket *sock, struct msghdr *m, size_t dlen)
 	rc = tipc_msg_build(hdr, m, 0, dlen, mtu, &pkts);
 	if (unlikely(rc != dlen))
 		return rc;
-	if (unlikely(syn && !tipc_msg_skb_clone(&pkts, &sk->sk_write_queue)))
+	if (unlikely(syn && !tipc_msg_skb_clone(&pkts, &sk->sk_write_queue))) {
+		__skb_queue_purge(&pkts);
 		return -ENOMEM;
+	}
 
 	trace_tipc_sk_sendmsg(sk, skb_peek(&pkts), TIPC_DUMP_SK_SNDQ, " ");
 	rc = tipc_node_xmit(net, &pkts, dnode, tsk->portid);
@@ -2757,6 +2762,7 @@ static void tipc_sk_timeout(struct timer_list *t)
 	if (sock_owned_by_user(sk)) {
 		sk_reset_timer(sk, &sk->sk_timer, jiffies + HZ / 20);
 		bh_unlock_sock(sk);
+		sock_put(sk);
 		return;
 	}
 

net/tls/tls_main.c

@@ -209,24 +209,15 @@ int tls_push_partial_record(struct sock *sk, struct tls_context *ctx,
 	return tls_push_sg(sk, ctx, sg, offset, flags);
 }
 
-bool tls_free_partial_record(struct sock *sk, struct tls_context *ctx)
+void tls_free_partial_record(struct sock *sk, struct tls_context *ctx)
 {
 	struct scatterlist *sg;
 
-	sg = ctx->partially_sent_record;
-	if (!sg)
-		return false;
-
-	while (1) {
+	for (sg = ctx->partially_sent_record; sg; sg = sg_next(sg)) {
 		put_page(sg_page(sg));
 		sk_mem_uncharge(sk, sg->length);
-
-		if (sg_is_last(sg))
-			break;
-		sg++;
 	}
 	ctx->partially_sent_record = NULL;
-	return true;
 }
 
 static void tls_write_space(struct sock *sk)

net/tls/tls_sw.c

@@ -710,8 +710,7 @@ static int tls_push_record(struct sock *sk, int flags,
 	}
 
 	i = msg_pl->sg.start;
-	sg_chain(rec->sg_aead_in, 2, rec->inplace_crypto ?
-		 &msg_en->sg.data[i] : &msg_pl->sg.data[i]);
+	sg_chain(rec->sg_aead_in, 2, &msg_pl->sg.data[i]);
 
 	i = msg_en->sg.end;
 	sk_msg_iter_var_prev(i);
@@ -771,8 +770,14 @@ static int bpf_exec_tx_verdict(struct sk_msg *msg, struct sock *sk,
 
 	policy = !(flags & MSG_SENDPAGE_NOPOLICY);
 	psock = sk_psock_get(sk);
-	if (!psock || !policy)
-		return tls_push_record(sk, flags, record_type);
+	if (!psock || !policy) {
+		err = tls_push_record(sk, flags, record_type);
+		if (err) {
+			*copied -= sk_msg_free(sk, msg);
+			tls_free_open_rec(sk);
+		}
+		return err;
+	}
 more_data:
 	enospc = sk_msg_full(msg);
 	if (psock->eval == __SK_NONE) {
@@ -970,8 +975,6 @@ int tls_sw_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)
 			if (ret)
 				goto fallback_to_reg_send;
 
-			rec->inplace_crypto = 0;
-
 			num_zc++;
 			copied += try_to_copy;
 
@@ -984,7 +987,7 @@ int tls_sw_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)
 					num_async++;
 				else if (ret == -ENOMEM)
 					goto wait_for_memory;
-				else if (ret == -ENOSPC)
+				else if (ctx->open_rec && ret == -ENOSPC)
 					goto rollback_iter;
 				else if (ret != -EAGAIN)
 					goto send_end;
@@ -1053,11 +1056,12 @@ int tls_sw_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)
 		ret = sk_stream_wait_memory(sk, &timeo);
 		if (ret) {
 trim_sgl:
-			tls_trim_both_msgs(sk, orig_size);
+			if (ctx->open_rec)
+				tls_trim_both_msgs(sk, orig_size);
 			goto send_end;
 		}
 
-		if (msg_en->sg.size < required_size)
+		if (ctx->open_rec && msg_en->sg.size < required_size)
 			goto alloc_encrypted;
 	}
 
@@ -1169,7 +1173,6 @@ static int tls_sw_do_sendpage(struct sock *sk, struct page *page,
 
 		tls_ctx->pending_open_record_frags = true;
 		if (full_record || eor || sk_msg_full(msg_pl)) {
-			rec->inplace_crypto = 0;
 			ret = bpf_exec_tx_verdict(msg_pl, sk, full_record,
 						  record_type, &copied, flags);
 			if (ret) {
@@ -1190,11 +1193,13 @@ static int tls_sw_do_sendpage(struct sock *sk, struct page *page,
 wait_for_memory:
 		ret = sk_stream_wait_memory(sk, &timeo);
 		if (ret) {
-			tls_trim_both_msgs(sk, msg_pl->sg.size);
+			if (ctx->open_rec)
+				tls_trim_both_msgs(sk, msg_pl->sg.size);
 			goto sendpage_end;
 		}
 
-		goto alloc_payload;
+		if (ctx->open_rec)
+			goto alloc_payload;
 	}
 
 	if (num_async) {
@@ -2084,7 +2089,8 @@ void tls_sw_release_resources_tx(struct sock *sk)
 	/* Free up un-sent records in tx_list. First, free
 	 * the partially sent record if any at head of tx_list.
 	 */
-	if (tls_free_partial_record(sk, tls_ctx)) {
+	if (tls_ctx->partially_sent_record) {
+		tls_free_partial_record(sk, tls_ctx);
 		rec = list_first_entry(&ctx->tx_list,
 				       struct tls_rec, list);
 		list_del(&rec->list);

tools/testing/selftests/bpf/test_sockmap.c

@@ -240,14 +240,14 @@ static int sockmap_init_sockets(int verbose)
 	addr.sin_port = htons(S1_PORT);
 	err = bind(s1, (struct sockaddr *)&addr, sizeof(addr));
 	if (err < 0) {
-		perror("bind s1 failed()\n");
+		perror("bind s1 failed()");
 		return errno;
 	}
 
 	addr.sin_port = htons(S2_PORT);
 	err = bind(s2, (struct sockaddr *)&addr, sizeof(addr));
 	if (err < 0) {
-		perror("bind s2 failed()\n");
+		perror("bind s2 failed()");
 		return errno;
 	}
 
@@ -255,14 +255,14 @@ static int sockmap_init_sockets(int verbose)
 	addr.sin_port = htons(S1_PORT);
 	err = listen(s1, 32);
 	if (err < 0) {
-		perror("listen s1 failed()\n");
+		perror("listen s1 failed()");
 		return errno;
 	}
 
 	addr.sin_port = htons(S2_PORT);
 	err = listen(s2, 32);
 	if (err < 0) {
-		perror("listen s1 failed()\n");
+		perror("listen s1 failed()");
 		return errno;
 	}
 
@@ -270,14 +270,14 @@ static int sockmap_init_sockets(int verbose)
 	addr.sin_port = htons(S1_PORT);
 	err = connect(c1, (struct sockaddr *)&addr, sizeof(addr));
 	if (err < 0 && errno != EINPROGRESS) {
-		perror("connect c1 failed()\n");
+		perror("connect c1 failed()");
 		return errno;
 	}
 
 	addr.sin_port = htons(S2_PORT);
 	err = connect(c2, (struct sockaddr *)&addr, sizeof(addr));
 	if (err < 0 && errno != EINPROGRESS) {
-		perror("connect c2 failed()\n");
+		perror("connect c2 failed()");
 		return errno;
 	} else if (err < 0) {
 		err = 0;
@@ -286,13 +286,13 @@ static int sockmap_init_sockets(int verbose)
 	/* Accept Connecrtions */
 	p1 = accept(s1, NULL, NULL);
 	if (p1 < 0) {
-		perror("accept s1 failed()\n");
+		perror("accept s1 failed()");
 		return errno;
 	}
 
 	p2 = accept(s2, NULL, NULL);
 	if (p2 < 0) {
-		perror("accept s1 failed()\n");
+		perror("accept s1 failed()");
 		return errno;
 	}
 
@@ -332,6 +332,10 @@ static int msg_loop_sendpage(int fd, int iov_length, int cnt,
 	int i, fp;
 
 	file = fopen(".sendpage_tst.tmp", "w+");
+	if (!file) {
+		perror("create file for sendpage");
+		return 1;
+	}
 	for (i = 0; i < iov_length * cnt; i++, k++)
 		fwrite(&k, sizeof(char), 1, file);
 	fflush(file);
@@ -339,12 +343,17 @@ static int msg_loop_sendpage(int fd, int iov_length, int cnt,
 	fclose(file);
 
 	fp = open(".sendpage_tst.tmp", O_RDONLY);
+	if (fp < 0) {
+		perror("reopen file for sendpage");
+		return 1;
+	}
+
 	clock_gettime(CLOCK_MONOTONIC, &s->start);
 	for (i = 0; i < cnt; i++) {
 		int sent = sendfile(fd, fp, NULL, iov_length);
 
 		if (!drop && sent < 0) {
-			perror("send loop error:");
+			perror("send loop error");
 			close(fp);
 			return sent;
 		} else if (drop && sent >= 0) {
@@ -463,7 +472,7 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
 			int sent = sendmsg(fd, &msg, flags);
 
 			if (!drop && sent < 0) {
-				perror("send loop error:");
+				perror("send loop error");
 				goto out_errno;
 			} else if (drop && sent >= 0) {
 				printf("send loop error expected: %i\n", sent);
@@ -499,7 +508,7 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
 		total_bytes -= txmsg_pop_total;
 		err = clock_gettime(CLOCK_MONOTONIC, &s->start);
 		if (err < 0)
-			perror("recv start time: ");
+			perror("recv start time");
 		while (s->bytes_recvd < total_bytes) {
 			if (txmsg_cork) {
 				timeout.tv_sec = 0;
@@ -543,7 +552,7 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
 			if (recv < 0) {
 				if (errno != EWOULDBLOCK) {
 					clock_gettime(CLOCK_MONOTONIC, &s->end);
-					perror("recv failed()\n");
+					perror("recv failed()");
 					goto out_errno;
 				}
 			}
@@ -557,7 +566,7 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
 
 				errno = msg_verify_data(&msg, recv, chunk_sz);
 				if (errno) {
-					perror("data verify msg failed\n");
+					perror("data verify msg failed");
 					goto out_errno;
 				}
 				if (recvp) {
@@ -565,7 +574,7 @@ static int msg_loop(int fd, int iov_count, int iov_length, int cnt,
 								recvp,
 								chunk_sz);
 					if (errno) {
-						perror("data verify msg_peek failed\n");
+						perror("data verify msg_peek failed");
 						goto out_errno;
 					}
 				}
@@ -654,7 +663,7 @@ static int sendmsg_test(struct sockmap_options *opt)
 			err = 0;
 		exit(err ? 1 : 0);
 	} else if (rxpid == -1) {
-		perror("msg_loop_rx: ");
+		perror("msg_loop_rx");
 		return errno;
 	}
 
@@ -681,7 +690,7 @@ static int sendmsg_test(struct sockmap_options *opt)
 				s.bytes_recvd, recvd_Bps, recvd_Bps/giga);
 		exit(err ? 1 : 0);
 	} else if (txpid == -1) {
-		perror("msg_loop_tx: ");
+		perror("msg_loop_tx");
 		return errno;
 	}
 
@@ -715,7 +724,7 @@ static int forever_ping_pong(int rate, struct sockmap_options *opt)
 	/* Ping/Pong data from client to server */
 	sc = send(c1, buf, sizeof(buf), 0);
 	if (sc < 0) {
-		perror("send failed()\n");
+		perror("send failed()");
 		return sc;
 	}
 
@@ -748,7 +757,7 @@ static int forever_ping_pong(int rate, struct sockmap_options *opt)
 			rc = recv(i, buf, sizeof(buf), 0);
 			if (rc < 0) {
 				if (errno != EWOULDBLOCK) {
-					perror("recv failed()\n");
+					perror("recv failed()");
 					return rc;
 				}
 			}
@@ -760,7 +769,7 @@ static int forever_ping_pong(int rate, struct sockmap_options *opt)
 
 			sc = send(i, buf, rc, 0);
 			if (sc < 0) {
-				perror("send failed()\n");
+				perror("send failed()");
 				return sc;
 			}
 		}

tools/testing/selftests/bpf/xdping.c

@@ -45,7 +45,7 @@ static int get_stats(int fd, __u16 count, __u32 raddr)
 	printf("\nXDP RTT data:\n");
 
 	if (bpf_map_lookup_elem(fd, &raddr, &pinginfo)) {
-		perror("bpf_map_lookup elem: ");
+		perror("bpf_map_lookup elem");
 		return 1;
 	}
 

tools/testing/selftests/net/forwarding/tc_common.sh

@@ -3,16 +3,42 @@
 
 CHECK_TC="yes"
 
+# Can be overridden by the configuration file. See lib.sh
+TC_HIT_TIMEOUT=${TC_HIT_TIMEOUT:=1000} # ms
+
+__tc_check_packets()
+{
+	local id=$1
+	local handle=$2
+	local count=$3
+	local operator=$4
+
+	start_time="$(date -u +%s%3N)"
+	while true
+	do
+		cmd_jq "tc -j -s filter show $id" \
+		       ".[] | select(.options.handle == $handle) | \
+			    select(.options.actions[0].stats.packets $operator $count)" \
+		    &> /dev/null
+		ret=$?
+		if [[ $ret -eq 0 ]]; then
+			return $ret
+		fi
+		current_time="$(date -u +%s%3N)"
+		diff=$(expr $current_time - $start_time)
+		if [ "$diff" -gt "$TC_HIT_TIMEOUT" ]; then
+			return 1
+		fi
+	done
+}
+
 tc_check_packets()
 {
 	local id=$1
 	local handle=$2
 	local count=$3
 
-	cmd_jq "tc -j -s filter show $id" \
-	       ".[] | select(.options.handle == $handle) | \
-	              select(.options.actions[0].stats.packets == $count)" \
-	       &> /dev/null
+	__tc_check_packets "$id" "$handle" "$count" "=="
 }
 
 tc_check_packets_hitting()
@@ -20,8 +46,5 @@ tc_check_packets_hitting()
 	local id=$1
 	local handle=$2
 
-	cmd_jq "tc -j -s filter show $id" \
-	       ".[] | select(.options.handle == $handle) | \
-		      select(.options.actions[0].stats.packets > 0)" \
-	       &> /dev/null
+	__tc_check_packets "$id" "$handle" 0 ">"
 }

tools/testing/selftests/net/pmtu.sh

@@ -1249,8 +1249,7 @@ test_list_flush_ipv4_exception() {
 	done
 	run_cmd ${ns_a} ping -q -M want -i 0.1 -c 2 -s 1800 "${dst2}"
 
-	# Each exception is printed as two lines
-	if [ "$(${ns_a} ip route list cache | wc -l)" -ne 202 ]; then
+	if [ "$(${ns_a} ip -oneline route list cache | wc -l)" -ne 101 ]; then
 		err "  can't list cached exceptions"
 		fail=1
 	fi
@@ -1300,7 +1299,7 @@ test_list_flush_ipv6_exception() {
 		run_cmd ${ns_a} ping -q -M want -i 0.1 -w 1 -s 1800 "${dst_prefix1}${i}"
 	done
 	run_cmd ${ns_a} ping -q -M want -i 0.1 -w 1 -s 1800 "${dst2}"
-	if [ "$(${ns_a} ip -6 route list cache | wc -l)" -ne 101 ]; then
+	if [ "$(${ns_a} ip -oneline -6 route list cache | wc -l)" -ne 101 ]; then
 		err "  can't list cached exceptions"
 		fail=1
 	fi

tools/testing/selftests/net/tls.c

@@ -268,6 +268,38 @@ TEST_F(tls, sendmsg_single)
 	EXPECT_EQ(memcmp(buf, test_str, send_len), 0);
 }
 
+#define MAX_FRAGS	64
+#define SEND_LEN	13
+TEST_F(tls, sendmsg_fragmented)
+{
+	char const *test_str = "test_sendmsg";
+	char buf[SEND_LEN * MAX_FRAGS];
+	struct iovec vec[MAX_FRAGS];
+	struct msghdr msg;
+	int i, frags;
+
+	for (frags = 1; frags <= MAX_FRAGS; frags++) {
+		for (i = 0; i < frags; i++) {
+			vec[i].iov_base = (char *)test_str;
+			vec[i].iov_len = SEND_LEN;
+		}
+
+		memset(&msg, 0, sizeof(struct msghdr));
+		msg.msg_iov = vec;
+		msg.msg_iovlen = frags;
+
+		EXPECT_EQ(sendmsg(self->fd, &msg, 0), SEND_LEN * frags);
+		EXPECT_EQ(recv(self->cfd, buf, SEND_LEN * frags, MSG_WAITALL),
+			  SEND_LEN * frags);
+
+		for (i = 0; i < frags; i++)
+			EXPECT_EQ(memcmp(buf + SEND_LEN * i,
+					 test_str, SEND_LEN), 0);
+	}
+}
+#undef MAX_FRAGS
+#undef SEND_LEN
+
 TEST_F(tls, sendmsg_large)
 {
 	void *mem = malloc(16384);
@@ -694,6 +726,34 @@ TEST_F(tls, recv_lowat)
 	EXPECT_EQ(memcmp(send_mem, recv_mem + 10, 5), 0);
 }
 
+TEST_F(tls, recv_rcvbuf)
+{
+	char send_mem[4096];
+	char recv_mem[4096];
+	int rcv_buf = 1024;
+
+	memset(send_mem, 0x1c, sizeof(send_mem));
+
+	EXPECT_EQ(setsockopt(self->cfd, SOL_SOCKET, SO_RCVBUF,
+			     &rcv_buf, sizeof(rcv_buf)), 0);
+
+	EXPECT_EQ(send(self->fd, send_mem, 512, 0), 512);
+	memset(recv_mem, 0, sizeof(recv_mem));
+	EXPECT_EQ(recv(self->cfd, recv_mem, sizeof(recv_mem), 0), 512);
+	EXPECT_EQ(memcmp(send_mem, recv_mem, 512), 0);
+
+	if (self->notls)
+		return;
+
+	EXPECT_EQ(send(self->fd, send_mem, 4096, 0), 4096);
+	memset(recv_mem, 0, sizeof(recv_mem));
+	EXPECT_EQ(recv(self->cfd, recv_mem, sizeof(recv_mem), 0), -1);
+	EXPECT_EQ(errno, EMSGSIZE);
+
+	EXPECT_EQ(recv(self->cfd, recv_mem, sizeof(recv_mem), 0), -1);
+	EXPECT_EQ(errno, EMSGSIZE);
+}
+
 TEST_F(tls, bidir)
 {
 	char const *test_str = "test_read";