Merge tag 'net-5.10-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

Pull networking fixes from Jakub Kicinski: "Cross-tree/merge window issues: - rtl8150: don't incorrectly assign random MAC addresses; fix late in the 5.9 cycle started depending on a return code from a function which changed with the 5.10 PR from the usb subsystem Current release regressions: - Revert "virtio-net: ethtool configurable RXCSUM", it was causing crashes at probe when control vq was not negotiated/available Previous release regressions: - ixgbe: fix probing of multi-port 10 Gigabit Intel NICs with an MDIO bus, only first device would be probed correctly - nexthop: Fix performance regression in nexthop deletion by effectively switching from recently added synchronize_rcu() to synchronize_rcu_expedited() - netsec: ignore 'phy-mode' device property on ACPI systems; the property is not populated correctly by the firmware, but firmware configures the PHY so just keep boot settings Previous releases - always broken: - tcp: fix to update snd_wl1 in bulk receiver fast path, addressing bulk transfers getting "stuck" - icmp: randomize the global rate limiter to prevent attackers from getting useful signal - r8169: fix operation under forced interrupt threading, make the driver always use hard irqs, even on RT, given the handler is light and only wants to schedule napi (and do so through a _irqoff() variant, preferably) - bpf: Enforce pointer id generation for all may-be-null register type to avoid pointers erroneously getting marked as null-checked - tipc: re-configure queue limit for broadcast link - net/sched: act_tunnel_key: fix OOB write in case of IPv6 ERSPAN tunnels - fix various issues in chelsio inline tls driver Misc: - bpf: improve just-added bpf_redirect_neigh() helper api to support supplying nexthop by the caller - in case BPF program has already done a lookup we can avoid doing another one - remove unnecessary break statements - make MCTCP not select IPV6, but rather depend on it" * tag 'net-5.10-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (62 commits) tcp: fix to update snd_wl1 in bulk receiver fast path net: Properly typecast int values to set sk_max_pacing_rate netfilter: nf_fwd_netdev: clear timestamp in forwarding path ibmvnic: save changed mac address to adapter->mac_addr selftests: mptcp: depends on built-in IPv6 Revert "virtio-net: ethtool configurable RXCSUM" rtnetlink: fix data overflow in rtnl_calcit() net: ethernet: mtk-star-emac: select REGMAP_MMIO net: hdlc_raw_eth: Clear the IFF_TX_SKB_SHARING flag after calling ether_setup net: hdlc: In hdlc_rcv, check to make sure dev is an HDLC device bpf, libbpf: Guard bpf inline asm from bpf_tail_call_static bpf, selftests: Extend test_tc_redirect to use modified bpf_redirect_neigh() bpf: Fix bpf_redirect_neigh helper api to support supplying nexthop mptcp: depends on IPV6 but not as a module sfc: move initialisation of efx->filter_sem to efx_init_struct() mpls: load mpls_gso after mpls_iptunnel net/sched: act_tunnel_key: fix OOB write in case of IPv6 ERSPAN tunnels net/sched: act_gate: Unlock ->tcfa_lock in tc_setup_flow_action() net: dsa: bcm_sf2: make const array static, makes object smaller mptcp: MPTCP_IPV6 should depend on IPV6 instead of selecting it ...

Linus Torvalds committed 6 years ago

Commit 3cb12d22 parents 0adc313 18ded9179 files · +676 −256Browse files

Unified · Split

79 files

Documentation/devicetree/bindings/net/socionext-netsec.txt +3 −1
Documentation/networking/ip-sysctl.rst +3 −1
Documentation/networking/nf_flowtable.rst +1 −1
MAINTAINERS +2 −1
drivers/net/dsa/bcm_sf2.c +1 −1
drivers/net/dsa/ocelot/seville_vsc9953.c +1 −1
drivers/net/ethernet/aquantia/atlantic/aq_nic.c +0 −1
drivers/net/ethernet/chelsio/inline_crypto/Kconfig +1 −0
drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c +13 −6
drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c +3 −2
drivers/net/ethernet/cisco/enic/enic_ethtool.c +0 −1
drivers/net/ethernet/faraday/ftgmac100.c +5 −0
drivers/net/ethernet/faraday/ftgmac100.h +8 −0
drivers/net/ethernet/ibm/ibmvnic.c +5 −0
drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c +14 −9
drivers/net/ethernet/intel/ixgbe/ixgbe_x540.c +0 −1
drivers/net/ethernet/korina.c +2 −2
drivers/net/ethernet/mediatek/Kconfig +1 −0
drivers/net/ethernet/realtek/r8169_main.c +4 −4
drivers/net/ethernet/sfc/efx_common.c +1 −0
drivers/net/ethernet/sfc/rx_common.c +0 −1
drivers/net/ethernet/socionext/netsec.c +17 −7
drivers/net/pcs/Kconfig +1 −2
drivers/net/usb/rtl8150.c +1 −1
drivers/net/virtio_net.c +12 −36
drivers/net/wan/hdlc.c +9 −1
drivers/net/wan/hdlc_raw_eth.c +1 −0
drivers/net/wan/lmc/lmc_proto.c +0 −4
drivers/nfc/st21nfca/core.c +0 −1
drivers/nfc/trf7970a.c +0 −1
include/linux/filter.h +9 −0
include/linux/netlink.h +1 −1
include/net/netfilter/nf_tables.h +6 −0
include/uapi/linux/bpf.h +18 −4
kernel/bpf/syscall.c +0 −1
kernel/bpf/verifier.c +5 −6
net/bridge/netfilter/ebt_dnat.c +1 −1
net/bridge/netfilter/ebt_redirect.c +1 −1
net/bridge/netfilter/ebt_snat.c +1 −1
net/core/dev.c +1 −1
net/core/filter.c +101 −60
net/core/rtnetlink.c +6 −7
net/core/sock.c +1 −1
net/dsa/tag_ksz.c +2 −0
net/ipv4/icmp.c +5 −2
net/ipv4/nexthop.c +1 −1
net/ipv4/tcp_input.c +2 −0
net/ipv6/netfilter/nf_conntrack_reasm.c +1 −0
net/mpls/mpls_iptunnel.c +1 −0
net/mptcp/Kconfig +2 −4
net/mptcp/options.c +2 −1
net/netfilter/ipvs/ip_vs_proto_tcp.c +6 −4
net/netfilter/nf_conntrack_proto_tcp.c +13 −6
net/netfilter/nf_dup_netdev.c +1 −0
net/netfilter/nf_tables_api.c +3 −3
net/netfilter/nf_tables_offload.c +2 −2
net/netfilter/nft_fwd_netdev.c +1 −0
net/nfc/netlink.c +1 −1
net/openvswitch/flow_table.c +35 −23
net/openvswitch/flow_table.h +6 −2
net/sched/act_ct.c +2 −2
net/sched/act_tunnel_key.c +1 −1
net/sched/cls_api.c +1 −1
net/tipc/bcast.c +8 −2
samples/bpf/sockex3_kern.c +4 −4
scripts/bpf_helpers_doc.py +1 −0
tools/include/uapi/linux/bpf.h +18 −4
tools/lib/bpf/bpf_helpers.h +2 −0
tools/testing/selftests/bpf/prog_tests/ksyms_btf.c +39 −18
tools/testing/selftests/bpf/progs/test_ksyms_btf_null_check.c +31 −0
tools/testing/selftests/bpf/progs/test_tc_neigh.c +3 −2
tools/testing/selftests/bpf/progs/test_tc_neigh_fib.c +155 −0
tools/testing/selftests/bpf/test_tc_redirect.sh +15 −3
tools/testing/selftests/bpf/verifier/sock.c +25 −0
tools/testing/selftests/net/config +1 −0
tools/testing/selftests/net/forwarding/vxlan_asymmetric.sh +10 −0
tools/testing/selftests/net/forwarding/vxlan_symmetric.sh +10 −0
tools/testing/selftests/net/mptcp/config +1 −0
tools/testing/selftests/net/rtnetlink.sh +5 −0

Documentation/devicetree/bindings/net/socionext-netsec.txt+3−1

@@ -30,7 +30,9 @@ Optional properties: (See ethernet.txt file in the same directory) · 29 unmodified lines
30	- max-frame-size: See ethernet.txt in the same directory.	30	- max-frame-size: See ethernet.txt in the same directory.
31		31
32	The MAC address will be determined using the optional properties	32	The MAC address will be determined using the optional properties
33	defined in ethernet.txt.	33	defined in ethernet.txt. The 'phy-mode' property is required, but may
		34	be set to the empty string if the PHY configuration is programmed by
		35	the firmware or set by hardware straps, and needs to be preserved.
34		36
35	Example:	37	Example:
36	eth0: ethernet@522d0000 {	38	eth0: ethernet@522d0000 {

Documentation/networking/ip-sysctl.rst+3−1

 icmp_msgs_per_sec - INTEGER
 	Limit maximal number of ICMP packets sent per second from this host.
 	Only messages whose type matches icmp_ratemask (see below) are
-	controlled by this limit.
+	controlled by this limit. For security reasons, the precise count
+	of messages per second is randomized.
 	Default: 1000
 icmp_msgs_burst - INTEGER
 	icmp_msgs_per_sec controls number of ICMP packets sent per second,
 	while icmp_msgs_burst controls the burst size of these packets.
+	For security reasons, the precise burst size is randomized.
 	Default: 50

Documentation/networking/nf_flowtable.rst+1−1

@@ -109,7 +109,7 @@ More reading · 108 unmodified lines
109	This documentation is based on the LWN.net articles [1]_\ [2]_. Rafal Milecki	109	This documentation is based on the LWN.net articles [1]_\ [2]_. Rafal Milecki
110	also made a very complete and comprehensive summary called "A state of network	110	also made a very complete and comprehensive summary called "A state of network
111	acceleration" that describes how things were before this infrastructure was	111	acceleration" that describes how things were before this infrastructure was
112	mailined [3]_ and it also makes a rough summary of this work [4]_.	112	mainlined [3]_ and it also makes a rough summary of this work [4]_.
113		113
114	.. [1] https://lwn.net/Articles/738214/	114	.. [1] https://lwn.net/Articles/738214/
115	.. [2] https://lwn.net/Articles/742164/	115	.. [2] https://lwn.net/Articles/742164/

MAINTAINERS+2−1

@@ -3244,7 +3244,8 @@ R: KP Singh <kpsingh@chromium.org> · 3243 unmodified lines
3244	L: netdev@vger.kernel.org	3244	L: netdev@vger.kernel.org
3245	L: bpf@vger.kernel.org	3245	L: bpf@vger.kernel.org
3246	S: Supported	3246	S: Supported
3247	Q: https://patchwork.ozlabs.org/project/netdev/list/?delegate=77147	3247	W: https://bpf.io/
		3248	Q: https://patchwork.kernel.org/project/netdevbpf/list/?delegate=121173
3248	T: git git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf.git	3249	T: git git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf.git
3249	T: git git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git	3250	T: git git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git
3250	F: Documentation/bpf/	3251	F: Documentation/bpf/

drivers/net/dsa/bcm_sf2.c+1−1

@@ -54,7 +54,7 @@ static void bcm_sf2_recalc_clock(struct dsa_switch *ds) · 53 unmodified lines
54	unsigned long new_rate;	54	unsigned long new_rate;
55	unsigned int ports_active;	55	unsigned int ports_active;
56	/* Frequenty in Mhz */	56	/* Frequenty in Mhz */
57	const unsigned long rate_table[] = {	57	static const unsigned long rate_table[] = {
58	59220000,	58	59220000,
59	60820000,	59	60820000,
60	62500000,	60	62500000,

drivers/net/dsa/ocelot/seville_vsc9953.c+1−1

@@ -1181,7 +1181,7 @@ static const struct felix_info seville_info_vsc9953 = { · 1180 unmodified lines
1181	.stats_layout = vsc9953_stats_layout,	1181	.stats_layout = vsc9953_stats_layout,
1182	.num_stats = ARRAY_SIZE(vsc9953_stats_layout),	1182	.num_stats = ARRAY_SIZE(vsc9953_stats_layout),
1183	.vcap = vsc9953_vcap_props,	1183	.vcap = vsc9953_vcap_props,
1184	.shared_queue_sz = 2048 * 1024,	1184	.shared_queue_sz = 256 * 1024,
1185	.num_mact_rows = 2048,	1185	.num_mact_rows = 2048,
1186	.num_ports = 10,	1186	.num_ports = 10,
1187	.mdio_bus_alloc = vsc9953_mdio_bus_alloc,	1187	.mdio_bus_alloc = vsc9953_mdio_bus_alloc,

drivers/net/ethernet/aquantia/atlantic/aq_nic.c+0−1

@@ -1163,7 +1163,6 @@ int aq_nic_set_link_ksettings(struct aq_nic_s *self, · 1162 unmodified lines
1163	default:	1163	default:
1164	err = -1;	1164	err = -1;
1165	goto err_exit;	1165	goto err_exit;
1166	break;
1167	}	1166	}
1168	if (!(self->aq_nic_cfg.aq_hw_caps->link_speed_msk & rate)) {	1167	if (!(self->aq_nic_cfg.aq_hw_caps->link_speed_msk & rate)) {
1169	err = -1;	1168	err = -1;

drivers/net/ethernet/chelsio/inline_crypto/Kconfig+1−0

@@ -16,6 +16,7 @@ if CHELSIO_INLINE_CRYPTO · 15 unmodified lines
16	config CRYPTO_DEV_CHELSIO_TLS	16	config CRYPTO_DEV_CHELSIO_TLS
17	tristate "Chelsio Crypto Inline TLS Driver"	17	tristate "Chelsio Crypto Inline TLS Driver"
18	depends on CHELSIO_T4	18	depends on CHELSIO_T4
		19	depends on TLS
19	depends on TLS_TOE	20	depends on TLS_TOE
20	help	21	help
21	Support Chelsio Inline TLS with Chelsio crypto accelerator.	22	Support Chelsio Inline TLS with Chelsio crypto accelerator.

drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c+13−6

@@ -92,11 +92,13 @@ static void chtls_sock_release(struct kref *ref) · 91 unmodified lines
92	static struct net_device chtls_find_netdev(struct chtls_dev cdev,	92	static struct net_device chtls_find_netdev(struct chtls_dev cdev,
93	struct sock *sk)	93	struct sock *sk)
94	{	94	{
		95	struct adapter *adap = pci_get_drvdata(cdev->pdev);
95	struct net_device *ndev = cdev->ports[0];	96	struct net_device *ndev = cdev->ports[0];
96	#if IS_ENABLED(CONFIG_IPV6)	97	#if IS_ENABLED(CONFIG_IPV6)
97	struct net_device *temp;	98	struct net_device *temp;
98	int addr_type;	99	int addr_type;
99	#endif	100	#endif
		101	int i;
100		102
101	switch (sk->sk_family) {	103	switch (sk->sk_family) {
102	case PF_INET:	104	case PF_INET:
@@ -127,8 +129,12 @@ static struct net_device chtls_find_netdev(struct chtls_dev cdev, · 24 unmodified lines
127	return NULL;	129	return NULL;
128		130
129	if (is_vlan_dev(ndev))	131	if (is_vlan_dev(ndev))
130	return vlan_dev_real_dev(ndev);	132	ndev = vlan_dev_real_dev(ndev);
131	return ndev;	133
		134	for_each_port(adap, i)
		135	if (cdev->ports[i] == ndev)
		136	return ndev;
		137	return NULL;
132	}	138	}
133		139
134	static void assign_rxopt(struct sock *sk, unsigned int opt)	140	static void assign_rxopt(struct sock *sk, unsigned int opt)
@@ -477,7 +483,6 @@ void chtls_destroy_sock(struct sock *sk) · 342 unmodified lines
477	chtls_purge_write_queue(sk);	483	chtls_purge_write_queue(sk);
478	free_tls_keyid(sk);	484	free_tls_keyid(sk);
479	kref_put(&csk->kref, chtls_sock_release);	485	kref_put(&csk->kref, chtls_sock_release);
480	csk->cdev = NULL;
481	if (sk->sk_family == AF_INET)	486	if (sk->sk_family == AF_INET)
482	sk->sk_prot = &tcp_prot;	487	sk->sk_prot = &tcp_prot;
483	#if IS_ENABLED(CONFIG_IPV6)	488	#if IS_ENABLED(CONFIG_IPV6)
@@ -736,14 +741,13 @@ void chtls_listen_stop(struct chtls_dev cdev, struct sock sk) · 252 unmodified lines
736		741
737	#if IS_ENABLED(CONFIG_IPV6)	742	#if IS_ENABLED(CONFIG_IPV6)
738	if (sk->sk_family == PF_INET6) {	743	if (sk->sk_family == PF_INET6) {
739	struct chtls_sock *csk;	744	struct net_device *ndev = chtls_find_netdev(cdev, sk);
740	int addr_type = 0;	745	int addr_type = 0;
741		746
742	csk = rcu_dereference_sk_user_data(sk);
743	addr_type = ipv6_addr_type((const struct in6_addr *)	747	addr_type = ipv6_addr_type((const struct in6_addr *)
744	&sk->sk_v6_rcv_saddr);	748	&sk->sk_v6_rcv_saddr);
745	if (addr_type != IPV6_ADDR_ANY)	749	if (addr_type != IPV6_ADDR_ANY)
746	cxgb4_clip_release(csk->egress_dev, (const u32 *)	750	cxgb4_clip_release(ndev, (const u32 *)
747	&sk->sk_v6_rcv_saddr, 1);	751	&sk->sk_v6_rcv_saddr, 1);
748	}	752	}
749	#endif	753	#endif
@@ -1157,6 +1161,9 @@ static struct sock chtls_recv_sock(struct sock lsk, · 407 unmodified lines
1157	ndev = n->dev;	1161	ndev = n->dev;
1158	if (!ndev)	1162	if (!ndev)
1159	goto free_dst;	1163	goto free_dst;
		1164	if (is_vlan_dev(ndev))
		1165	ndev = vlan_dev_real_dev(ndev);
		1166
1160	port_id = cxgb4_port_idx(ndev);	1167	port_id = cxgb4_port_idx(ndev);
1161		1168
1162	csk = chtls_sock_create(cdev);	1169	csk = chtls_sock_create(cdev);

drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_io.c+3−2

 	return 0;
 }
-static int csk_mem_free(struct chtls_dev *cdev, struct sock *sk)
+static bool csk_mem_free(struct chtls_dev *cdev, struct sock *sk)
 {
-	return (cdev->max_host_sndbuf - sk->sk_wmem_queued);
+	return (cdev->max_host_sndbuf - sk->sk_wmem_queued > 0);
 }
 static int csk_wait_memory(struct chtls_dev *cdev,
 	copied = 0;
 	csk = rcu_dereference_sk_user_data(sk);
 	cdev = csk->cdev;
+	lock_sock(sk);
 	timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT);
 	err = sk_stream_wait_connect(sk, &timeo);

drivers/net/ethernet/cisco/enic/enic_ethtool.c+0−1

@@ -434,7 +434,6 @@ static int enic_grxclsrule(struct enic enic, struct ethtool_rxnfc cmd) · 433 unmodified lines
434	break;	434	break;
435	default:	435	default:
436	return -EINVAL;	436	return -EINVAL;
437	break;
438	}	437	}
439		438
440	fsp->h_u.tcp_ip4_spec.ip4src = flow_get_u32_src(&n->keys);	439	fsp->h_u.tcp_ip4_spec.ip4src = flow_get_u32_src(&n->keys);

drivers/net/ethernet/faraday/ftgmac100.c+5−0

 		priv->rxdes0_edorr_mask = BIT(30);
 		priv->txdes0_edotr_mask = BIT(30);
 		priv->is_aspeed = true;
+		/* Disable ast2600 problematic HW arbitration */
+		if (of_device_is_compatible(np, "aspeed,ast2600-mac")) {
+			iowrite32(FTGMAC100_TM_DEFAULT,
+				  priv->base + FTGMAC100_OFFSET_TM);
+		}
 	} else {
 		priv->rxdes0_edorr_mask = BIT(15);
 		priv->txdes0_edotr_mask = BIT(15);

drivers/net/ethernet/faraday/ftgmac100.h+8−0

@@ -169,6 +169,14 @@ · 168 unmodified lines
169	#define FTGMAC100_MACCR_FAST_MODE (1 << 19)	169	#define FTGMAC100_MACCR_FAST_MODE (1 << 19)
170	#define FTGMAC100_MACCR_SW_RST (1 << 31)	170	#define FTGMAC100_MACCR_SW_RST (1 << 31)
171		171
		172	/*
		173	* test mode control register
		174	*/
		175	#define FTGMAC100_TM_RQ_TX_VALID_DIS (1 << 28)
		176	#define FTGMAC100_TM_RQ_RR_IDLE_PREV (1 << 27)
		177	#define FTGMAC100_TM_DEFAULT \
		178	(FTGMAC100_TM_RQ_TX_VALID_DIS \| FTGMAC100_TM_RQ_RR_IDLE_PREV)
		179
172	/*	180	/*
173	* PHY control register	181	* PHY control register
174	*/	182	*/

drivers/net/ethernet/ibm/ibmvnic.c+5−0

 		dev_err(dev, "Error %ld in CHANGE_MAC_ADDR_RSP\n", rc);
 		goto out;
 	}
+	/* crq->change_mac_addr.mac_addr is the requested one
+	 * crq->change_mac_addr_rsp.mac_addr is the returned valid one.
+	 */
 	ether_addr_copy(netdev->dev_addr,
 			&crq->change_mac_addr_rsp.mac_addr[0]);
+	ether_addr_copy(adapter->mac_addr,
+			&crq->change_mac_addr_rsp.mac_addr[0]);
 out:
 	complete(&adapter->fw_done);
 	return rc;

drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c+14−9

  **/
 s32 ixgbe_mii_bus_init(struct ixgbe_hw *hw)
 {
+	s32 (*write)(struct mii_bus *bus, int addr, int regnum, u16 val);
+	s32 (*read)(struct mii_bus *bus, int addr, int regnum);
 	struct ixgbe_adapter *adapter = hw->back;
 	struct pci_dev *pdev = adapter->pdev;
 	struct device *dev = &adapter->netdev->dev;
 	struct mii_bus *bus;
-	bus = devm_mdiobus_alloc(dev);
-	if (!bus)
-		return -ENOMEM;
 	switch (hw->device_id) {
 	/* C3000 SoCs */
 	case IXGBE_DEV_ID_X550EM_A_KR:
 	case IXGBE_DEV_ID_X550EM_A_1G_T:
 	case IXGBE_DEV_ID_X550EM_A_1G_T_L:
 		if (!ixgbe_x550em_a_has_mii(hw))
-			return -ENODEV;
+			return 0;
-		bus->read = &ixgbe_x550em_a_mii_bus_read;
+		read = &ixgbe_x550em_a_mii_bus_read;
-		bus->write = &ixgbe_x550em_a_mii_bus_write;
+		write = &ixgbe_x550em_a_mii_bus_write;
 		break;
 	default:
-		bus->read = &ixgbe_mii_bus_read;
+		read = &ixgbe_mii_bus_read;
-		bus->write = &ixgbe_mii_bus_write;
+		write = &ixgbe_mii_bus_write;
 		break;
 	}
+	bus = devm_mdiobus_alloc(dev);
+	if (!bus)
+		return -ENOMEM;
+	bus->read = read;
+	bus->write = write;
 	/* Use the position of the device in the PCI hierarchy as the id */
 	snprintf(bus->id, MII_BUS_ID_SIZE, "%s-mdio-%s", ixgbe_driver_name,
 		 pci_name(pdev));

drivers/net/ethernet/intel/ixgbe/ixgbe_x540.c+0−1

@@ -350,7 +350,6 @@ static s32 ixgbe_calc_eeprom_checksum_X540(struct ixgbe_hw *hw) · 349 unmodified lines
350	if (ixgbe_read_eerd_generic(hw, pointer, &length)) {	350	if (ixgbe_read_eerd_generic(hw, pointer, &length)) {
351	hw_dbg(hw, "EEPROM read failed\n");	351	hw_dbg(hw, "EEPROM read failed\n");
352	return IXGBE_ERR_EEPROM;	352	return IXGBE_ERR_EEPROM;
353	break;
354	}	353	}
355		354
356	/* Skip pointer section if length is invalid. */	355	/* Skip pointer section if length is invalid. */

drivers/net/ethernet/korina.c+2−2

@@ -1113,7 +1113,7 @@ out: · 1112 unmodified lines
1113	return rc;	1113	return rc;
1114		1114
1115	probe_err_register:	1115	probe_err_register:
1116	kfree(KSEG0ADDR(lp->td_ring));	1116	kfree((struct dma_desc *)KSEG0ADDR(lp->td_ring));
1117	probe_err_td_ring:	1117	probe_err_td_ring:
1118	iounmap(lp->tx_dma_regs);	1118	iounmap(lp->tx_dma_regs);
1119	probe_err_dma_tx:	1119	probe_err_dma_tx:
@@ -1133,7 +1133,7 @@ static int korina_remove(struct platform_device *pdev) · 13 unmodified lines
1133	iounmap(lp->eth_regs);	1133	iounmap(lp->eth_regs);
1134	iounmap(lp->rx_dma_regs);	1134	iounmap(lp->rx_dma_regs);
1135	iounmap(lp->tx_dma_regs);	1135	iounmap(lp->tx_dma_regs);
1136	kfree(KSEG0ADDR(lp->td_ring));	1136	kfree((struct dma_desc *)KSEG0ADDR(lp->td_ring));
1137		1137
1138	unregister_netdev(bif->dev);	1138	unregister_netdev(bif->dev);
1139	free_netdev(bif->dev);	1139	free_netdev(bif->dev);

drivers/net/ethernet/mediatek/Kconfig+1−0

@@ -17,6 +17,7 @@ config NET_MEDIATEK_SOC · 16 unmodified lines
17	config NET_MEDIATEK_STAR_EMAC	17	config NET_MEDIATEK_STAR_EMAC
18	tristate "MediaTek STAR Ethernet MAC support"	18	tristate "MediaTek STAR Ethernet MAC support"
19	select PHYLIB	19	select PHYLIB
		20	select REGMAP_MMIO
20	help	21	help
21	This driver supports the ethernet MAC IP first used on	22	This driver supports the ethernet MAC IP first used on
22	MediaTek MT85** SoCs.	23	MediaTek MT85** SoCs.

drivers/net/ethernet/realtek/r8169_main.c+4−4

@@ -4694,7 +4694,7 @@ static int rtl8169_close(struct net_device *dev) · 4693 unmodified lines
4694		4694
4695	phy_disconnect(tp->phydev);	4695	phy_disconnect(tp->phydev);
4696		4696
4697	pci_free_irq(pdev, 0, tp);	4697	free_irq(pci_irq_vector(pdev, 0), tp);
4698		4698
4699	dma_free_coherent(&pdev->dev, R8169_RX_RING_BYTES, tp->RxDescArray,	4699	dma_free_coherent(&pdev->dev, R8169_RX_RING_BYTES, tp->RxDescArray,
4700	tp->RxPhyAddr);	4700	tp->RxPhyAddr);
@@ -4745,8 +4745,8 @@ static int rtl_open(struct net_device *dev) · 44 unmodified lines
4745		4745
4746	rtl_request_firmware(tp);	4746	rtl_request_firmware(tp);
4747		4747
4748	retval = pci_request_irq(pdev, 0, rtl8169_interrupt, NULL, tp,	4748	retval = request_irq(pci_irq_vector(pdev, 0), rtl8169_interrupt,
4749	dev->name);	4749	IRQF_NO_THREAD \| IRQF_SHARED, dev->name, tp);
4750	if (retval < 0)	4750	if (retval < 0)
4751	goto err_release_fw_2;	4751	goto err_release_fw_2;
4752		4752
@@ -4763,7 +4763,7 @@ out: · 10 unmodified lines
4763	return retval;	4763	return retval;
4764		4764
4765	err_free_irq:	4765	err_free_irq:
4766	pci_free_irq(pdev, 0, tp);	4766	free_irq(pci_irq_vector(pdev, 0), tp);
4767	err_release_fw_2:	4767	err_release_fw_2:
4768	rtl_release_firmware(tp);	4768	rtl_release_firmware(tp);
4769	rtl8169_rx_clear(tp);	4769	rtl8169_rx_clear(tp);

drivers/net/ethernet/sfc/efx_common.c+1−0

@@ -1014,6 +1014,7 @@ int efx_init_struct(struct efx_nic *efx, · 1013 unmodified lines
1014	efx->num_mac_stats = MC_CMD_MAC_NSTATS;	1014	efx->num_mac_stats = MC_CMD_MAC_NSTATS;
1015	BUILD_BUG_ON(MC_CMD_MAC_NSTATS - 1 != MC_CMD_MAC_GENERATION_END);	1015	BUILD_BUG_ON(MC_CMD_MAC_NSTATS - 1 != MC_CMD_MAC_GENERATION_END);
1016	mutex_init(&efx->mac_lock);	1016	mutex_init(&efx->mac_lock);
		1017	init_rwsem(&efx->filter_sem);
1017	#ifdef CONFIG_RFS_ACCEL	1018	#ifdef CONFIG_RFS_ACCEL
1018	mutex_init(&efx->rps_mutex);	1019	mutex_init(&efx->rps_mutex);
1019	spin_lock_init(&efx->rps_hash_lock);	1020	spin_lock_init(&efx->rps_hash_lock);

drivers/net/ethernet/sfc/rx_common.c+0−1

@@ -797,7 +797,6 @@ int efx_probe_filters(struct efx_nic *efx) · 796 unmodified lines
797	{	797	{
798	int rc;	798	int rc;
799		799
800	init_rwsem(&efx->filter_sem);
801	mutex_lock(&efx->mac_lock);	800	mutex_lock(&efx->mac_lock);
802	down_write(&efx->filter_sem);	801	down_write(&efx->filter_sem);
803	rc = efx->type->filter_table_probe(efx);	802	rc = efx->type->filter_table_probe(efx);

drivers/net/ethernet/socionext/netsec.c+17−7

 #include <linux/pm_runtime.h>
 #include <linux/acpi.h>
 #include <linux/of_mdio.h>
+#include <linux/of_net.h>
 #include <linux/etherdevice.h>
 #include <linux/interrupt.h>
 #include <linux/io.h>
 static int netsec_of_probe(struct platform_device *pdev,
 			   struct netsec_priv *priv, u32 *phy_addr)
 {
+	int err;
+	err = of_get_phy_mode(pdev->dev.of_node, &priv->phy_interface);
+	if (err) {
+		dev_err(&pdev->dev, "missing required property 'phy-mode'\n");
+		return err;
+	}
 	priv->phy_np = of_parse_phandle(pdev->dev.of_node, "phy-handle", 0);
 	if (!priv->phy_np) {
 		dev_err(&pdev->dev, "missing required property 'phy-handle'\n");
 	if (!IS_ENABLED(CONFIG_ACPI))
 		return -ENODEV;
+	/* ACPI systems are assumed to configure the PHY in firmware, so
+	 * there is really no need to discover the PHY mode from the DSDT.
+	 * Since firmware is known to exist in the field that configures the
+	 * PHY correctly but passes the wrong mode string in the phy-mode
+	 * device property, we have no choice but to ignore it.
+	 */
+	priv->phy_interface = PHY_INTERFACE_MODE_NA;
 	ret = device_property_read_u32(&pdev->dev, "phy-channel", phy_addr);
 	if (ret) {
 		dev_err(&pdev->dev,
 	priv->msg_enable = NETIF_MSG_TX_ERR | NETIF_MSG_HW | NETIF_MSG_DRV |
 			   NETIF_MSG_LINK | NETIF_MSG_PROBE;
-	priv->phy_interface = device_get_phy_mode(&pdev->dev);
-	if ((int)priv->phy_interface < 0) {
-		dev_err(&pdev->dev, "missing required property 'phy-mode'\n");
-		ret = -ENODEV;
-		goto free_ndev;
-	}
 	priv->ioaddr = devm_ioremap(&pdev->dev, mmio_res->start,
 				    resource_size(mmio_res));
 	if (!priv->ioaddr) {

drivers/net/pcs/Kconfig+1−2

@@ -7,8 +7,7 @@ menu "PCS device drivers" · 6 unmodified lines
7		7
8	config PCS_XPCS	8	config PCS_XPCS
9	tristate "Synopsys DesignWare XPCS controller"	9	tristate "Synopsys DesignWare XPCS controller"
10	select MDIO_BUS	10	depends on MDIO_DEVICE && MDIO_BUS
11	depends on MDIO_DEVICE
12	help	11	help
13	This module provides helper functions for Synopsys DesignWare XPCS	12	This module provides helper functions for Synopsys DesignWare XPCS
14	controllers.	13	controllers.

drivers/net/usb/rtl8150.c+1−1

@@ -261,7 +261,7 @@ static void set_ethernet_addr(rtl8150_t *dev) · 260 unmodified lines
261		261
262	ret = get_registers(dev, IDR, sizeof(node_id), node_id);	262	ret = get_registers(dev, IDR, sizeof(node_id), node_id);
263		263
264	if (ret == sizeof(node_id)) {	264	if (!ret) {
265	ether_addr_copy(dev->netdev->dev_addr, node_id);	265	ether_addr_copy(dev->netdev->dev_addr, node_id);
266	} else {	266	} else {
267	eth_hw_addr_random(dev->netdev);	267	eth_hw_addr_random(dev->netdev);

drivers/net/virtio_net.c+12−36

 				(1ULL << VIRTIO_NET_F_GUEST_ECN)  | \
 				(1ULL << VIRTIO_NET_F_GUEST_UFO))
-#define GUEST_OFFLOAD_CSUM_MASK (1ULL << VIRTIO_NET_F_GUEST_CSUM)
 struct virtnet_stat_desc {
 	char desc[ETH_GSTRING_LEN];
 	size_t offset;
 	return 0;
 }
-static netdev_features_t virtnet_fix_features(struct net_device *netdev,
-					      netdev_features_t features)
-{
-	/* If Rx checksum is disabled, LRO should also be disabled. */
-	if (!(features & NETIF_F_RXCSUM))
-		features &= ~NETIF_F_LRO;
-	return features;
-}
 static int virtnet_set_features(struct net_device *dev,
 				netdev_features_t features)
 {
 	struct virtnet_info *vi = netdev_priv(dev);
-	u64 offloads = vi->guest_offloads;
+	u64 offloads;
 	int err;
-	/* Don't allow configuration while XDP is active. */
-	if (vi->xdp_queue_pairs)
-		return -EBUSY;
 	if ((dev->features ^ features) & NETIF_F_LRO) {
+		if (vi->xdp_queue_pairs)
+			return -EBUSY;
 		if (features & NETIF_F_LRO)
-			offloads |= GUEST_OFFLOAD_LRO_MASK &
+			offloads = vi->guest_offloads_capable;
-				    vi->guest_offloads_capable;
 		else
-			offloads &= ~GUEST_OFFLOAD_LRO_MASK;
+			offloads = vi->guest_offloads_capable &
-	}
+				   ~GUEST_OFFLOAD_LRO_MASK;
-	if ((dev->features ^ features) & NETIF_F_RXCSUM) {
+		err = virtnet_set_guest_offloads(vi, offloads);
-		if (features & NETIF_F_RXCSUM)
+		if (err)
-			offloads |= GUEST_OFFLOAD_CSUM_MASK &
+			return err;
-				    vi->guest_offloads_capable;
+		vi->guest_offloads = offloads;
-		else
-			offloads &= ~GUEST_OFFLOAD_CSUM_MASK;
 	}
-	err = virtnet_set_guest_offloads(vi, offloads);
-	if (err)
-		return err;
-	vi->guest_offloads = offloads;
 	return 0;
 }
 	.ndo_features_check	= passthru_features_check,
 	.ndo_get_phys_port_name	= virtnet_get_phys_port_name,
 	.ndo_set_features	= virtnet_set_features,
-	.ndo_fix_features	= virtnet_fix_features,
 };
 static void virtnet_config_changed_work(struct work_struct *work)
 	if (virtio_has_feature(vdev, VIRTIO_NET_F_GUEST_TSO4) ||
 	    virtio_has_feature(vdev, VIRTIO_NET_F_GUEST_TSO6))
 		dev->features |= NETIF_F_LRO;
-	if (virtio_has_feature(vdev, VIRTIO_NET_F_CTRL_GUEST_OFFLOADS)) {
+	if (virtio_has_feature(vdev, VIRTIO_NET_F_CTRL_GUEST_OFFLOADS))
-		dev->hw_features |= NETIF_F_RXCSUM;
 		dev->hw_features |= NETIF_F_LRO;
-	}
 	dev->vlan_features = dev->features;

drivers/net/wan/hdlc.c+9−1

@@ -46,7 +46,15 @@ static struct hdlc_proto *first_proto; · 45 unmodified lines
46	static int hdlc_rcv(struct sk_buff skb, struct net_device dev,	46	static int hdlc_rcv(struct sk_buff skb, struct net_device dev,
47	struct packet_type p, struct net_device orig_dev)	47	struct packet_type p, struct net_device orig_dev)
48	{	48	{
49	struct hdlc_device *hdlc = dev_to_hdlc(dev);	49	struct hdlc_device *hdlc;
		50
		51	/* First make sure "dev" is an HDLC device */
		52	if (!(dev->priv_flags & IFF_WAN_HDLC)) {
		53	kfree_skb(skb);
		54	return NET_RX_SUCCESS;
		55	}
		56
		57	hdlc = dev_to_hdlc(dev);
50		58
51	if (!net_eq(dev_net(dev), &init_net)) {	59	if (!net_eq(dev_net(dev), &init_net)) {
52	kfree_skb(skb);	60	kfree_skb(skb);

drivers/net/wan/hdlc_raw_eth.c+1−0

@@ -99,6 +99,7 @@ static int raw_eth_ioctl(struct net_device dev, struct ifreq ifr) · 98 unmodified lines
99	old_qlen = dev->tx_queue_len;	99	old_qlen = dev->tx_queue_len;
100	ether_setup(dev);	100	ether_setup(dev);
101	dev->tx_queue_len = old_qlen;	101	dev->tx_queue_len = old_qlen;
		102	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
102	eth_hw_addr_random(dev);	103	eth_hw_addr_random(dev);
103	call_netdevice_notifiers(NETDEV_POST_TYPE_CHANGE, dev);	104	call_netdevice_notifiers(NETDEV_POST_TYPE_CHANGE, dev);
104	netif_dormant_off(dev);	105	netif_dormant_off(dev);

drivers/net/wan/lmc/lmc_proto.c+0−4

     switch(sc->if_type){
     case LMC_PPP:
 	    return hdlc_type_trans(skb, sc->lmc_device);
-	    break;
     case LMC_NET:
         return htons(ETH_P_802_2);
-        break;
     case LMC_RAW: /* Packet type for skbuff kind of useless */
         return htons(ETH_P_802_2);
-        break;
     default:
         printk(KERN_WARNING "%s: No protocol set for this interface, assuming 802.2 (which is wrong!!)\n", sc->name);
         return htons(ETH_P_802_2);
-        break;
     }
 }

drivers/nfc/st21nfca/core.c+0−1

@@ -794,7 +794,6 @@ static int st21nfca_hci_im_transceive(struct nfc_hci_dev *hdev, · 793 unmodified lines
794	skb->len,	794	skb->len,
795	st21nfca_hci_data_exchange_cb,	795	st21nfca_hci_data_exchange_cb,
796	info);	796	info);
797	break;
798	default:	797	default:
799	return 1;	798	return 1;
800	}	799	}

drivers/nfc/trf7970a.c+0−1

@@ -1382,7 +1382,6 @@ static int trf7970a_is_iso15693_write_or_lock(u8 cmd) · 1381 unmodified lines
1382	case ISO15693_CMD_WRITE_DSFID:	1382	case ISO15693_CMD_WRITE_DSFID:
1383	case ISO15693_CMD_LOCK_DSFID:	1383	case ISO15693_CMD_LOCK_DSFID:
1384	return 1;	1384	return 1;
1385	break;
1386	default:	1385	default:
1387	return 0;	1386	return 0;
1388	}	1387	}

include/linux/filter.h+9−0

 	void *data_end;
 };
+struct bpf_nh_params {
+	u32 nh_family;
+	union {
+		u32 ipv4_nh;
+		struct in6_addr ipv6_nh;
+	};
+};
 struct bpf_redirect_info {
 	u32 flags;
 	u32 tgt_index;
 	void *tgt_value;
 	struct bpf_map *map;
 	u32 kern_flags;
+	struct bpf_nh_params nh;
 };
 DECLARE_PER_CPU(struct bpf_redirect_info, bpf_redirect_info);

include/linux/netlink.h+1−1

@@ -240,7 +240,7 @@ struct netlink_dump_control { · 239 unmodified lines
240	int (done)(struct netlink_callback );	240	int (done)(struct netlink_callback );
241	void *data;	241	void *data;
242	struct module *module;	242	struct module *module;
243	u16 min_dump_alloc;	243	u32 min_dump_alloc;
244	};	244	};
245		245
246	int __netlink_dump_start(struct sock ssk, struct sk_buff skb,	246	int __netlink_dump_start(struct sock ssk, struct sk_buff skb,

include/net/netfilter/nf_tables.h+6−0

 	return (struct nft_expr *)&rule->data[rule->dlen];
 }
+static inline bool nft_expr_more(const struct nft_rule *rule,
+				 const struct nft_expr *expr)
+{
+	return expr != nft_expr_last(rule) && expr->ops;
+}
 static inline struct nft_userdata *nft_userdata(const struct nft_rule *rule)
 {
 	return (void *)&rule->data[rule->dlen];

include/uapi/linux/bpf.h+18−4

  * 	Return
  * 		The id is returned or 0 in case the id could not be retrieved.
  *
- * long bpf_redirect_neigh(u32 ifindex, u64 flags)
+ * long bpf_redirect_neigh(u32 ifindex, struct bpf_redir_neigh *params, int plen, u64 flags)
  * 	Description
  * 		Redirect the packet to another net device of index *ifindex*
  * 		and fill in L2 addresses from neighboring subsystem. This helper
  * 		is somewhat similar to **bpf_redirect**\ (), except that it
  * 		populates L2 addresses as well, meaning, internally, the helper
- * 		performs a FIB lookup based on the skb's networking header to
+ * 		relies on the neighbor lookup for the L2 address of the nexthop.
- * 		get the address of the next hop and then relies on the neighbor
+ *
- * 		lookup for the L2 address of the nexthop.
+ * 		The helper will perform a FIB lookup based on the skb's
+ * 		networking header to get the address of the next hop, unless
+ * 		this is supplied by the caller in the *params* argument. The
+ * 		*plen* argument indicates the len of *params* and should be set
+ * 		to 0 if *params* is NULL.
  *
  * 		The *flags* argument is reserved and must be 0. The helper is
  * 		currently only supported for tc BPF program types, and enabled
 	__u8	dmac[6];     /* ETH_ALEN */
 };
+struct bpf_redir_neigh {
+	/* network family for lookup (AF_INET, AF_INET6) */
+	__u32 nh_family;
+	/* network address of nexthop; skips fib lookup to find gateway */
+	union {
+		__be32		ipv4_nh;
+		__u32		ipv6_nh[4];  /* in6_addr; network order */
+	};
+};
 enum bpf_task_fd_type {
 	BPF_FD_TYPE_RAW_TRACEPOINT,	/* tp name */
 	BPF_FD_TYPE_TRACEPOINT,		/* tp name */

kernel/bpf/syscall.c+0−1

@@ -2913,7 +2913,6 @@ attach_type_to_prog_type(enum bpf_attach_type attach_type) · 2912 unmodified lines
2913	case BPF_CGROUP_INET_INGRESS:	2913	case BPF_CGROUP_INET_INGRESS:
2914	case BPF_CGROUP_INET_EGRESS:	2914	case BPF_CGROUP_INET_EGRESS:
2915	return BPF_PROG_TYPE_CGROUP_SKB;	2915	return BPF_PROG_TYPE_CGROUP_SKB;
2916	break;
2917	case BPF_CGROUP_INET_SOCK_CREATE:	2916	case BPF_CGROUP_INET_SOCK_CREATE:
2918	case BPF_CGROUP_INET_SOCK_RELEASE:	2917	case BPF_CGROUP_INET_SOCK_RELEASE:
2919	case BPF_CGROUP_INET4_POST_BIND:	2918	case BPF_CGROUP_INET4_POST_BIND:

kernel/bpf/verifier.c+5−6

 				regs[BPF_REG_0].id = ++env->id_gen;
 		} else {
 			regs[BPF_REG_0].type = PTR_TO_MAP_VALUE_OR_NULL;
-			regs[BPF_REG_0].id = ++env->id_gen;
 		}
 	} else if (fn->ret_type == RET_PTR_TO_SOCKET_OR_NULL) {
 		mark_reg_known_zero(env, regs, BPF_REG_0);
 		regs[BPF_REG_0].type = PTR_TO_SOCKET_OR_NULL;
-		regs[BPF_REG_0].id = ++env->id_gen;
 	} else if (fn->ret_type == RET_PTR_TO_SOCK_COMMON_OR_NULL) {
 		mark_reg_known_zero(env, regs, BPF_REG_0);
 		regs[BPF_REG_0].type = PTR_TO_SOCK_COMMON_OR_NULL;
-		regs[BPF_REG_0].id = ++env->id_gen;
 	} else if (fn->ret_type == RET_PTR_TO_TCP_SOCK_OR_NULL) {
 		mark_reg_known_zero(env, regs, BPF_REG_0);
 		regs[BPF_REG_0].type = PTR_TO_TCP_SOCK_OR_NULL;
-		regs[BPF_REG_0].id = ++env->id_gen;
 	} else if (fn->ret_type == RET_PTR_TO_ALLOC_MEM_OR_NULL) {
 		mark_reg_known_zero(env, regs, BPF_REG_0);
 		regs[BPF_REG_0].type = PTR_TO_MEM_OR_NULL;
-		regs[BPF_REG_0].id = ++env->id_gen;
 		regs[BPF_REG_0].mem_size = meta.mem_size;
 	} else if (fn->ret_type == RET_PTR_TO_MEM_OR_BTF_ID_OR_NULL ||
 		   fn->ret_type == RET_PTR_TO_MEM_OR_BTF_ID) {
 		return -EINVAL;
 	}
+	if (reg_type_may_be_null(regs[BPF_REG_0].type))
+		regs[BPF_REG_0].id = ++env->id_gen;
 	if (is_ptr_cast_function(func_id)) {
 		/* For release_reference() */
 		regs[BPF_REG_0].ref_obj_id = meta.ref_obj_id;
 				 struct bpf_reg_state *reg, u32 id,
 				 bool is_null)
 {
-	if (reg_type_may_be_null(reg->type) && reg->id == id) {
+	if (reg_type_may_be_null(reg->type) && reg->id == id &&
+	    !WARN_ON_ONCE(!reg->id)) {
 		/* Old offset (both fixed and variable parts) should
 		 * have been known-zero, because we don't allow pointer
 		 * arithmetic on pointers that might be NULL.

net/bridge/netfilter/ebt_dnat.c+1−1

@@ -21,7 +21,7 @@ ebt_dnat_tg(struct sk_buff skb, const struct xt_action_param par) · 20 unmodified lines
21	{	21	{
22	const struct ebt_nat_info *info = par->targinfo;	22	const struct ebt_nat_info *info = par->targinfo;
23		23
24	if (skb_ensure_writable(skb, ETH_ALEN))	24	if (skb_ensure_writable(skb, 0))
25	return EBT_DROP;	25	return EBT_DROP;
26		26
27	ether_addr_copy(eth_hdr(skb)->h_dest, info->mac);	27	ether_addr_copy(eth_hdr(skb)->h_dest, info->mac);

net/bridge/netfilter/ebt_redirect.c+1−1

@@ -21,7 +21,7 @@ ebt_redirect_tg(struct sk_buff skb, const struct xt_action_param par) · 20 unmodified lines
21	{	21	{
22	const struct ebt_redirect_info *info = par->targinfo;	22	const struct ebt_redirect_info *info = par->targinfo;
23		23
24	if (skb_ensure_writable(skb, ETH_ALEN))	24	if (skb_ensure_writable(skb, 0))
25	return EBT_DROP;	25	return EBT_DROP;
26		26
27	if (xt_hooknum(par) != NF_BR_BROUTING)	27	if (xt_hooknum(par) != NF_BR_BROUTING)

net/bridge/netfilter/ebt_snat.c+1−1

@@ -22,7 +22,7 @@ ebt_snat_tg(struct sk_buff skb, const struct xt_action_param par) · 21 unmodified lines
22	{	22	{
23	const struct ebt_nat_info *info = par->targinfo;	23	const struct ebt_nat_info *info = par->targinfo;
24		24
25	if (skb_ensure_writable(skb, ETH_ALEN * 2))	25	if (skb_ensure_writable(skb, 0))
26	return EBT_DROP;	26	return EBT_DROP;
27		27
28	ether_addr_copy(eth_hdr(skb)->h_source, info->mac);	28	ether_addr_copy(eth_hdr(skb)->h_source, info->mac);

net/core/dev.c+1−1

@@ -10213,7 +10213,7 @@ void netdev_run_todo(void) · 10212 unmodified lines
10213	struct net_device *dev = list_first_entry(&unlink_list,	10213	struct net_device *dev = list_first_entry(&unlink_list,
10214	struct net_device,	10214	struct net_device,
10215	unlink_list);	10215	unlink_list);
10216	list_del(&dev->unlink_list);	10216	list_del_init(&dev->unlink_list);
10217	dev->nested_level = dev->lower_level - 1;	10217	dev->nested_level = dev->lower_level - 1;
10218	}	10218	}
10219	#endif	10219	#endif

net/core/filter.c+101−60

 }
 #if IS_ENABLED(CONFIG_IPV6)
-static int bpf_out_neigh_v6(struct net *net, struct sk_buff *skb)
+static int bpf_out_neigh_v6(struct net *net, struct sk_buff *skb,
+			    struct net_device *dev, struct bpf_nh_params *nh)
 {
-	struct dst_entry *dst = skb_dst(skb);
-	struct net_device *dev = dst->dev;
 	u32 hh_len = LL_RESERVED_SPACE(dev);
 	const struct in6_addr *nexthop;
+	struct dst_entry *dst = NULL;
 	struct neighbour *neigh;
 	if (dev_xmit_recursion()) {
 	}
 	rcu_read_lock_bh();
-	nexthop = rt6_nexthop(container_of(dst, struct rt6_info, dst),
+	if (!nh) {
-			      &ipv6_hdr(skb)->daddr);
+		dst = skb_dst(skb);
+		nexthop = rt6_nexthop(container_of(dst, struct rt6_info, dst),
+				      &ipv6_hdr(skb)->daddr);
+	} else {
+		nexthop = &nh->ipv6_nh;
+	}
 	neigh = ip_neigh_gw6(dev, nexthop);
 	if (likely(!IS_ERR(neigh))) {
 		int ret;
 		return ret;
 	}
 	rcu_read_unlock_bh();
-	IP6_INC_STATS(dev_net(dst->dev),
+	if (dst)
-		      ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES);
+		IP6_INC_STATS(dev_net(dst->dev),
+			      ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES);
 out_drop:
 	kfree_skb(skb);
 	return -ENETDOWN;
 }
-static int __bpf_redirect_neigh_v6(struct sk_buff *skb, struct net_device *dev)
+static int __bpf_redirect_neigh_v6(struct sk_buff *skb, struct net_device *dev,
+				   struct bpf_nh_params *nh)
 {
 	const struct ipv6hdr *ip6h = ipv6_hdr(skb);
 	struct net *net = dev_net(dev);
 	int err, ret = NET_XMIT_DROP;
-	struct dst_entry *dst;
-	struct flowi6 fl6 = {
-		.flowi6_flags	= FLOWI_FLAG_ANYSRC,
-		.flowi6_mark	= skb->mark,
-		.flowlabel	= ip6_flowinfo(ip6h),
-		.flowi6_oif	= dev->ifindex,
-		.flowi6_proto	= ip6h->nexthdr,
-		.daddr		= ip6h->daddr,
-		.saddr		= ip6h->saddr,
-	};
-	dst = ipv6_stub->ipv6_dst_lookup_flow(net, NULL, &fl6, NULL);
+	if (!nh) {
-	if (IS_ERR(dst))
+		struct dst_entry *dst;
-		goto out_drop;
+		struct flowi6 fl6 = {
+			.flowi6_flags = FLOWI_FLAG_ANYSRC,
+			.flowi6_mark  = skb->mark,
+			.flowlabel    = ip6_flowinfo(ip6h),
+			.flowi6_oif   = dev->ifindex,
+			.flowi6_proto = ip6h->nexthdr,
+			.daddr	      = ip6h->daddr,
+			.saddr	      = ip6h->saddr,
+		};
+		dst = ipv6_stub->ipv6_dst_lookup_flow(net, NULL, &fl6, NULL);
+		if (IS_ERR(dst))
+			goto out_drop;
-	skb_dst_set(skb, dst);
+		skb_dst_set(skb, dst);
+	} else if (nh->nh_family != AF_INET6) {
+		goto out_drop;
+	}
-	err = bpf_out_neigh_v6(net, skb);
+	err = bpf_out_neigh_v6(net, skb, dev, nh);
 	if (unlikely(net_xmit_eval(err)))
 		dev->stats.tx_errors++;
 	else
 	return ret;
 }
 #else
-static int __bpf_redirect_neigh_v6(struct sk_buff *skb, struct net_device *dev)
+static int __bpf_redirect_neigh_v6(struct sk_buff *skb, struct net_device *dev,
+				   struct bpf_nh_params *nh)
 {
 	kfree_skb(skb);
 	return NET_XMIT_DROP;
 #endif /* CONFIG_IPV6 */
 #if IS_ENABLED(CONFIG_INET)
-static int bpf_out_neigh_v4(struct net *net, struct sk_buff *skb)
+static int bpf_out_neigh_v4(struct net *net, struct sk_buff *skb,
+			    struct net_device *dev, struct bpf_nh_params *nh)
 {
-	struct dst_entry *dst = skb_dst(skb);
-	struct rtable *rt = container_of(dst, struct rtable, dst);
-	struct net_device *dev = dst->dev;
 	u32 hh_len = LL_RESERVED_SPACE(dev);
 	struct neighbour *neigh;
 	bool is_v6gw = false;
 	}
 	rcu_read_lock_bh();
-	neigh = ip_neigh_for_gw(rt, skb, &is_v6gw);
+	if (!nh) {
+		struct dst_entry *dst = skb_dst(skb);
+		struct rtable *rt = container_of(dst, struct rtable, dst);
+		neigh = ip_neigh_for_gw(rt, skb, &is_v6gw);
+	} else if (nh->nh_family == AF_INET6) {
+		neigh = ip_neigh_gw6(dev, &nh->ipv6_nh);
+		is_v6gw = true;
+	} else if (nh->nh_family == AF_INET) {
+		neigh = ip_neigh_gw4(dev, nh->ipv4_nh);
+	} else {
+		rcu_read_unlock_bh();
+		goto out_drop;
+	}
 	if (likely(!IS_ERR(neigh))) {
 		int ret;
 	return -ENETDOWN;
 }
-static int __bpf_redirect_neigh_v4(struct sk_buff *skb, struct net_device *dev)
+static int __bpf_redirect_neigh_v4(struct sk_buff *skb, struct net_device *dev,
+				   struct bpf_nh_params *nh)
 {
 	const struct iphdr *ip4h = ip_hdr(skb);
 	struct net *net = dev_net(dev);
 	int err, ret = NET_XMIT_DROP;
-	struct rtable *rt;
-	struct flowi4 fl4 = {
-		.flowi4_flags	= FLOWI_FLAG_ANYSRC,
-		.flowi4_mark	= skb->mark,
-		.flowi4_tos	= RT_TOS(ip4h->tos),
-		.flowi4_oif	= dev->ifindex,
-		.flowi4_proto	= ip4h->protocol,
-		.daddr		= ip4h->daddr,
-		.saddr		= ip4h->saddr,
-	};
-	rt = ip_route_output_flow(net, &fl4, NULL);
+	if (!nh) {
-	if (IS_ERR(rt))
+		struct flowi4 fl4 = {
-		goto out_drop;
+			.flowi4_flags = FLOWI_FLAG_ANYSRC,
-	if (rt->rt_type != RTN_UNICAST && rt->rt_type != RTN_LOCAL) {
+			.flowi4_mark  = skb->mark,
-		ip_rt_put(rt);
+			.flowi4_tos   = RT_TOS(ip4h->tos),
-		goto out_drop;
+			.flowi4_oif   = dev->ifindex,
-	}
+			.flowi4_proto = ip4h->protocol,
+			.daddr	      = ip4h->daddr,
+			.saddr	      = ip4h->saddr,
+		};
+		struct rtable *rt;
+		rt = ip_route_output_flow(net, &fl4, NULL);
+		if (IS_ERR(rt))
+			goto out_drop;
+		if (rt->rt_type != RTN_UNICAST && rt->rt_type != RTN_LOCAL) {
+			ip_rt_put(rt);
+			goto out_drop;
+		}
-	skb_dst_set(skb, &rt->dst);
+		skb_dst_set(skb, &rt->dst);
+	}
-	err = bpf_out_neigh_v4(net, skb);
+	err = bpf_out_neigh_v4(net, skb, dev, nh);
 	if (unlikely(net_xmit_eval(err)))
 		dev->stats.tx_errors++;
 	else
 	return ret;
 }
 #else
-static int __bpf_redirect_neigh_v4(struct sk_buff *skb, struct net_device *dev)
+static int __bpf_redirect_neigh_v4(struct sk_buff *skb, struct net_device *dev,
+				   struct bpf_nh_params *nh)
 {
 	kfree_skb(skb);
 	return NET_XMIT_DROP;
 }
 #endif /* CONFIG_INET */
-static int __bpf_redirect_neigh(struct sk_buff *skb, struct net_device *dev)
+static int __bpf_redirect_neigh(struct sk_buff *skb, struct net_device *dev,
+				struct bpf_nh_params *nh)
 {
 	struct ethhdr *ethh = eth_hdr(skb);
 	skb_reset_network_header(skb);
 	if (skb->protocol == htons(ETH_P_IP))
-		return __bpf_redirect_neigh_v4(skb, dev);
+		return __bpf_redirect_neigh_v4(skb, dev, nh);
 	else if (skb->protocol == htons(ETH_P_IPV6))
-		return __bpf_redirect_neigh_v6(skb, dev);
+		return __bpf_redirect_neigh_v6(skb, dev, nh);
 out:
 	kfree_skb(skb);
 	return -ENOTSUPP;
 enum {
 	BPF_F_NEIGH	= (1ULL << 1),
 	BPF_F_PEER	= (1ULL << 2),
-#define BPF_F_REDIRECT_INTERNAL	(BPF_F_NEIGH | BPF_F_PEER)
+	BPF_F_NEXTHOP	= (1ULL << 3),
+#define BPF_F_REDIRECT_INTERNAL	(BPF_F_NEIGH | BPF_F_PEER | BPF_F_NEXTHOP)
 };
 BPF_CALL_3(bpf_clone_redirect, struct sk_buff *, skb, u32, ifindex, u64, flags)
 		return -EAGAIN;
 	}
 	return flags & BPF_F_NEIGH ?
-	       __bpf_redirect_neigh(skb, dev) :
+	       __bpf_redirect_neigh(skb, dev, flags & BPF_F_NEXTHOP ?
+				    &ri->nh : NULL) :
 	       __bpf_redirect(skb, dev, flags);
 out_drop:
 	kfree_skb(skb);
 	.arg2_type      = ARG_ANYTHING,
 };
-BPF_CALL_2(bpf_redirect_neigh, u32, ifindex, u64, flags)
+BPF_CALL_4(bpf_redirect_neigh, u32, ifindex, struct bpf_redir_neigh *, params,
+	   int, plen, u64, flags)
 {
 	struct bpf_redirect_info *ri = this_cpu_ptr(&bpf_redirect_info);
-	if (unlikely(flags))
+	if (unlikely((plen && plen < sizeof(*params)) || flags))
 		return TC_ACT_SHOT;
-	ri->flags = BPF_F_NEIGH;
+	ri->flags = BPF_F_NEIGH | (plen ? BPF_F_NEXTHOP : 0);
 	ri->tgt_index = ifindex;
+	BUILD_BUG_ON(sizeof(struct bpf_redir_neigh) != sizeof(struct bpf_nh_params));
+	if (plen)
+		memcpy(&ri->nh, params, sizeof(ri->nh));
 	return TC_ACT_REDIRECT;
 }
 	.gpl_only	= false,
 	.ret_type	= RET_INTEGER,
 	.arg1_type	= ARG_ANYTHING,
-	.arg2_type	= ARG_ANYTHING,
+	.arg2_type      = ARG_PTR_TO_MEM_OR_NULL,
+	.arg3_type      = ARG_CONST_SIZE_OR_ZERO,
+	.arg4_type	= ARG_ANYTHING,
 };
 BPF_CALL_2(bpf_msg_apply_bytes, struct sk_msg *, msg, u32, bytes)
 				cmpxchg(&sk->sk_pacing_status,
 					SK_PACING_NONE,
 					SK_PACING_NEEDED);
-			sk->sk_max_pacing_rate = (val == ~0U) ? ~0UL : val;
+			sk->sk_max_pacing_rate = (val == ~0U) ?
+						 ~0UL : (unsigned int)val;
 			sk->sk_pacing_rate = min(sk->sk_pacing_rate,
 						 sk->sk_max_pacing_rate);
 			break;

net/core/rtnetlink.c+6−7

 	return rtnl_linkprop(RTM_DELLINKPROP, skb, nlh, extack);
 }
-static u16 rtnl_calcit(struct sk_buff *skb, struct nlmsghdr *nlh)
+static u32 rtnl_calcit(struct sk_buff *skb, struct nlmsghdr *nlh)
 {
 	struct net *net = sock_net(skb->sk);
-	struct net_device *dev;
+	size_t min_ifinfo_dump_size = 0;
 	struct nlattr *tb[IFLA_MAX+1];
 	u32 ext_filter_mask = 0;
-	u16 min_ifinfo_dump_size = 0;
+	struct net_device *dev;
 	int hdrlen;
 	/* Same kernel<->userspace interface hack as in rtnl_dump_ifinfo. */
 	 */
 	rcu_read_lock();
 	for_each_netdev_rcu(net, dev) {
-		min_ifinfo_dump_size = max_t(u16, min_ifinfo_dump_size,
+		min_ifinfo_dump_size = max(min_ifinfo_dump_size,
-					     if_nlmsg_size(dev,
+					   if_nlmsg_size(dev, ext_filter_mask));
-						           ext_filter_mask));
 	}
 	rcu_read_unlock();
 	if (kind == 2 && nlh->nlmsg_flags&NLM_F_DUMP) {
 		struct sock *rtnl;
 		rtnl_dumpit_func dumpit;
-		u16 min_dump_alloc = 0;
+		u32 min_dump_alloc = 0;
 		link = rtnl_get_link(family, type);
 		if (!link || !link->dumpit) {

net/core/sock.c+1−1

@@ -1163,7 +1163,7 @@ set_sndbuf: · 1162 unmodified lines
1163		1163
1164	case SO_MAX_PACING_RATE:	1164	case SO_MAX_PACING_RATE:
1165	{	1165	{
1166	unsigned long ulval = (val == ~0U) ? ~0UL : val;	1166	unsigned long ulval = (val == ~0U) ? ~0UL : (unsigned int)val;
1167		1167
1168	if (sizeof(ulval) != sizeof(val) &&	1168	if (sizeof(ulval) != sizeof(val) &&
1169	optlen >= sizeof(ulval) &&	1169	optlen >= sizeof(ulval) &&

net/dsa/tag_ksz.c+2−0

 	.xmit	= ksz8795_xmit,
 	.rcv	= ksz8795_rcv,
 	.overhead = KSZ_INGRESS_TAG_LEN,
+	.tail_tag = true,
 };
 DSA_TAG_DRIVER(ksz8795_netdev_ops);
 	.xmit	= ksz9477_xmit,
 	.rcv	= ksz9477_rcv,
 	.overhead = KSZ9477_INGRESS_TAG_LEN,
+	.tail_tag = true,
 };
 DSA_TAG_DRIVER(ksz9477_netdev_ops);

net/ipv4/icmp.c+5−2

 /**
  * icmp_global_allow - Are we allowed to send one more ICMP message ?
  *
- * Uses a token bucket to limit our ICMP messages to sysctl_icmp_msgs_per_sec.
+ * Uses a token bucket to limit our ICMP messages to ~sysctl_icmp_msgs_per_sec.
  * Returns false if we reached the limit and can not send another packet.
  * Note: called with BH disabled
  */
 	}
 	credit = min_t(u32, icmp_global.credit + incr, sysctl_icmp_msgs_burst);
 	if (credit) {
-		credit--;
+		/* We want to use a credit of one in average, but need to randomize
+		 * it for security reasons.
+		 */
+		credit = max_t(int, credit - prandom_u32_max(3), 0);
 		rc = true;
 	}
 	WRITE_ONCE(icmp_global.credit, credit);

net/ipv4/nexthop.c+1−1

@@ -845,7 +845,7 @@ static void remove_nexthop_from_groups(struct net net, struct nexthop nh, · 844 unmodified lines
845	remove_nh_grp_entry(net, nhge, nlinfo);	845	remove_nh_grp_entry(net, nhge, nlinfo);
846		846
847	/* make sure all see the newly published array before releasing rtnl */	847	/* make sure all see the newly published array before releasing rtnl */
848	synchronize_rcu();	848	synchronize_net();
849	}	849	}
850		850
851	static void remove_nexthop_group(struct nexthop nh, struct nl_info nlinfo)	851	static void remove_nexthop_group(struct nexthop nh, struct nl_info nlinfo)

net/ipv4/tcp_input.c+2−0

@@ -5827,6 +5827,8 @@ void tcp_rcv_established(struct sock sk, struct sk_buff skb) · 5826 unmodified lines
5827	tcp_data_snd_check(sk);	5827	tcp_data_snd_check(sk);
5828	if (!inet_csk_ack_scheduled(sk))	5828	if (!inet_csk_ack_scheduled(sk))
5829	goto no_ack;	5829	goto no_ack;
		5830	} else {
		5831	tcp_update_wl(tp, TCP_SKB_CB(skb)->seq);
5830	}	5832	}
5831		5833
5832	__tcp_ack_snd_check(sk, 0);	5834	__tcp_ack_snd_check(sk, 0);

net/ipv6/netfilter/nf_conntrack_reasm.c+1−0

@@ -355,6 +355,7 @@ static int nf_ct_frag6_reasm(struct frag_queue fq, struct sk_buff skb, · 354 unmodified lines
355	ipv6_hdr(skb)->payload_len = htons(payload_len);	355	ipv6_hdr(skb)->payload_len = htons(payload_len);
356	ipv6_change_dsfield(ipv6_hdr(skb), 0xff, ecn);	356	ipv6_change_dsfield(ipv6_hdr(skb), 0xff, ecn);
357	IP6CB(skb)->frag_max_size = sizeof(struct ipv6hdr) + fq->q.max_size;	357	IP6CB(skb)->frag_max_size = sizeof(struct ipv6hdr) + fq->q.max_size;
		358	IP6CB(skb)->flags \|= IP6SKB_FRAGMENTED;
358		359
359	/* Yes, and fold redundant checksum back. 8) */	360	/* Yes, and fold redundant checksum back. 8) */
360	if (skb->ip_summed == CHECKSUM_COMPLETE)	361	if (skb->ip_summed == CHECKSUM_COMPLETE)

net/mpls/mpls_iptunnel.c+1−0

@@ -300,5 +300,6 @@ static void __exit mpls_iptunnel_exit(void) · 299 unmodified lines
300	module_exit(mpls_iptunnel_exit);	300	module_exit(mpls_iptunnel_exit);
301		301
302	MODULE_ALIAS_RTNL_LWT(MPLS);	302	MODULE_ALIAS_RTNL_LWT(MPLS);
		303	MODULE_SOFTDEP("post: mpls_gso");
303	MODULE_DESCRIPTION("MultiProtocol Label Switching IP Tunnels");	304	MODULE_DESCRIPTION("MultiProtocol Label Switching IP Tunnels");
304	MODULE_LICENSE("GPL v2");	305	MODULE_LICENSE("GPL v2");

net/mptcp/Kconfig+2−4

 config MPTCP_IPV6
 	bool "MPTCP: IPv6 support for Multipath TCP"
-	select IPV6
+	depends on IPV6=y
 	default y
-endif
 config MPTCP_KUNIT_TESTS
 	tristate "This builds the MPTCP KUnit tests" if !KUNIT_ALL_TESTS
-	select MPTCP
 	depends on KUNIT
 	default KUNIT_ALL_TESTS
 	help
 	  If unsure, say N.
+endif

net/mptcp/options.c+2−1

 		}
 		mp_opt->add_addr = 1;
-		mp_opt->port = 0;
 		mp_opt->addr_id = *ptr++;
 		pr_debug("ADD_ADDR: id=%d, echo=%d", mp_opt->addr_id, mp_opt->echo);
 		if (mp_opt->family == MPTCP_ADDR_IPVERSION_4) {
 	mp_opt->mp_capable = 0;
 	mp_opt->mp_join = 0;
 	mp_opt->add_addr = 0;
+	mp_opt->ahmac = 0;
+	mp_opt->port = 0;
 	mp_opt->rm_addr = 0;
 	mp_opt->dss = 0;

net/netfilter/ipvs/ip_vs_proto_tcp.c+6−4

 	if (new_state != cp->state) {
 		struct ip_vs_dest *dest = cp->dest;
-		IP_VS_DBG_BUF(8, "%s %s [%c%c%c%c] %s:%d->"
+		IP_VS_DBG_BUF(8, "%s %s [%c%c%c%c] c:%s:%d v:%s:%d "
-			      "%s:%d state: %s->%s conn->refcnt:%d\n",
+			      "d:%s:%d state: %s->%s conn->refcnt:%d\n",
 			      pd->pp->name,
 			      ((state_off == TCP_DIR_OUTPUT) ?
 			       "output " : "input "),
 			      th->fin ? 'F' : '.',
 			      th->ack ? 'A' : '.',
 			      th->rst ? 'R' : '.',
-			      IP_VS_DBG_ADDR(cp->daf, &cp->daddr),
-			      ntohs(cp->dport),
 			      IP_VS_DBG_ADDR(cp->af, &cp->caddr),
 			      ntohs(cp->cport),
+			      IP_VS_DBG_ADDR(cp->af, &cp->vaddr),
+			      ntohs(cp->vport),
+			      IP_VS_DBG_ADDR(cp->daf, &cp->daddr),
+			      ntohs(cp->dport),
 			      tcp_state_name(cp->state),
 			      tcp_state_name(new_state),
 			      refcount_read(&cp->refcnt));

net/netfilter/nf_conntrack_proto_tcp.c+13−6

 			swin = win << sender->td_scale;
 			sender->td_maxwin = (swin == 0 ? 1 : swin);
 			sender->td_maxend = end + sender->td_maxwin;
-			/*
+			if (receiver->td_maxwin == 0) {
-			 * We haven't seen traffic in the other direction yet
+				/* We haven't seen traffic in the other
-			 * but we have to tweak window tracking to pass III
+				 * direction yet but we have to tweak window
-			 * and IV until that happens.
+				 * tracking to pass III and IV until that
-			 */
+				 * happens.
-			if (receiver->td_maxwin == 0)
+				 */
 				receiver->td_end = receiver->td_maxend = sack;
+			} else if (sack == receiver->td_end + 1) {
+				/* Likely a reply to a keepalive.
+				 * Needed for III.
+				 */
+				receiver->td_end++;
+			}
 		}
 	} else if (((state->state == TCP_CONNTRACK_SYN_SENT
 		     && dir == IP_CT_DIR_ORIGINAL)

net/netfilter/nf_dup_netdev.c+1−0

@@ -19,6 +19,7 @@ static void nf_do_netdev_egress(struct sk_buff skb, struct net_device dev) · 18 unmodified lines
19	skb_push(skb, skb->mac_len);	19	skb_push(skb, skb->mac_len);
20		20
21	skb->dev = dev;	21	skb->dev = dev;
		22	skb->tstamp = 0;
22	dev_queue_xmit(skb);	23	dev_queue_xmit(skb);
23	}	24	}
24		25

net/netfilter/nf_tables_api.c+3−3

@@ -302,7 +302,7 @@ static void nft_rule_expr_activate(const struct nft_ctx *ctx, · 301 unmodified lines
302	struct nft_expr *expr;	302	struct nft_expr *expr;
303		303
304	expr = nft_expr_first(rule);	304	expr = nft_expr_first(rule);
305	while (expr != nft_expr_last(rule) && expr->ops) {	305	while (nft_expr_more(rule, expr)) {
306	if (expr->ops->activate)	306	if (expr->ops->activate)
307	expr->ops->activate(ctx, expr);	307	expr->ops->activate(ctx, expr);
308		308
@@ -317,7 +317,7 @@ static void nft_rule_expr_deactivate(const struct nft_ctx *ctx, · 8 unmodified lines
317	struct nft_expr *expr;	317	struct nft_expr *expr;
318		318
319	expr = nft_expr_first(rule);	319	expr = nft_expr_first(rule);
320	while (expr != nft_expr_last(rule) && expr->ops) {	320	while (nft_expr_more(rule, expr)) {
321	if (expr->ops->deactivate)	321	if (expr->ops->deactivate)
322	expr->ops->deactivate(ctx, expr, phase);	322	expr->ops->deactivate(ctx, expr, phase);
323		323
@@ -3080,7 +3080,7 @@ static void nf_tables_rule_destroy(const struct nft_ctx *ctx, · 2756 unmodified lines
3080	* is called on error from nf_tables_newrule().	3080	* is called on error from nf_tables_newrule().
3081	*/	3081	*/
3082	expr = nft_expr_first(rule);	3082	expr = nft_expr_first(rule);
3083	while (expr != nft_expr_last(rule) && expr->ops) {	3083	while (nft_expr_more(rule, expr)) {
3084	next = nft_expr_next(expr);	3084	next = nft_expr_next(expr);
3085	nf_tables_expr_destroy(ctx, expr);	3085	nf_tables_expr_destroy(ctx, expr);
3086	expr = next;	3086	expr = next;

net/netfilter/nf_tables_offload.c+2−2

@@ -37,7 +37,7 @@ struct nft_flow_rule nft_flow_rule_create(struct net net, · 36 unmodified lines
37	struct nft_expr *expr;	37	struct nft_expr *expr;
38		38
39	expr = nft_expr_first(rule);	39	expr = nft_expr_first(rule);
40	while (expr->ops && expr != nft_expr_last(rule)) {	40	while (nft_expr_more(rule, expr)) {
41	if (expr->ops->offload_flags & NFT_OFFLOAD_F_ACTION)	41	if (expr->ops->offload_flags & NFT_OFFLOAD_F_ACTION)
42	num_actions++;	42	num_actions++;
43		43
@@ -61,7 +61,7 @@ struct nft_flow_rule nft_flow_rule_create(struct net net, · 17 unmodified lines
61	ctx->net = net;	61	ctx->net = net;
62	ctx->dep.type = NFT_OFFLOAD_DEP_UNSPEC;	62	ctx->dep.type = NFT_OFFLOAD_DEP_UNSPEC;
63		63
64	while (expr->ops && expr != nft_expr_last(rule)) {	64	while (nft_expr_more(rule, expr)) {
65	if (!expr->ops->offload) {	65	if (!expr->ops->offload) {
66	err = -EOPNOTSUPP;	66	err = -EOPNOTSUPP;
67	goto err_out;	67	goto err_out;

net/netfilter/nft_fwd_netdev.c+1−0

@@ -138,6 +138,7 @@ static void nft_fwd_neigh_eval(const struct nft_expr *expr, · 137 unmodified lines
138	return;	138	return;
139		139
140	skb->dev = dev;	140	skb->dev = dev;
		141	skb->tstamp = 0;
141	neigh_xmit(neigh_table, dev, addr, skb);	142	neigh_xmit(neigh_table, dev, addr, skb);
142	out:	143	out:
143	regs->verdict.code = verdict;	144	regs->verdict.code = verdict;

net/nfc/netlink.c+1−1

@@ -1217,7 +1217,7 @@ static int nfc_genl_fw_download(struct sk_buff skb, struct genl_info info) · 1216 unmodified lines
1217	u32 idx;	1217	u32 idx;
1218	char firmware_name[NFC_FIRMWARE_NAME_MAXSIZE + 1];	1218	char firmware_name[NFC_FIRMWARE_NAME_MAXSIZE + 1];
1219		1219
1220	if (!info->attrs[NFC_ATTR_DEVICE_INDEX])	1220	if (!info->attrs[NFC_ATTR_DEVICE_INDEX] \|\| !info->attrs[NFC_ATTR_FIRMWARE_NAME])
1221	return -EINVAL;	1221	return -EINVAL;
1222		1222
1223	idx = nla_get_u32(info->attrs[NFC_ATTR_DEVICE_INDEX]);	1223	idx = nla_get_u32(info->attrs[NFC_ATTR_DEVICE_INDEX]);

net/openvswitch/flow_table.c+35−23

 static void __mask_array_destroy(struct mask_array *ma)
 {
-	free_percpu(ma->masks_usage_cntr);
+	free_percpu(ma->masks_usage_stats);
 	kfree(ma);
 }
 		ma->masks_usage_zero_cntr[i] = 0;
 		for_each_possible_cpu(cpu) {
-			u64 *usage_counters = per_cpu_ptr(ma->masks_usage_cntr,
+			struct mask_array_stats *stats;
-							  cpu);
 			unsigned int start;
 			u64 counter;
+			stats = per_cpu_ptr(ma->masks_usage_stats, cpu);
 			do {
-				start = u64_stats_fetch_begin_irq(&ma->syncp);
+				start = u64_stats_fetch_begin_irq(&stats->syncp);
-				counter = usage_counters[i];
+				counter = stats->usage_cntrs[i];
-			} while (u64_stats_fetch_retry_irq(&ma->syncp, start));
+			} while (u64_stats_fetch_retry_irq(&stats->syncp, start));
 			ma->masks_usage_zero_cntr[i] += counter;
 		}
 					     sizeof(struct sw_flow_mask *) *
 					     size);
-	new->masks_usage_cntr = __alloc_percpu(sizeof(u64) * size,
+	new->masks_usage_stats = __alloc_percpu(sizeof(struct mask_array_stats) +
-					       __alignof__(u64));
+						sizeof(u64) * size,
-	if (!new->masks_usage_cntr) {
+						__alignof__(u64));
+	if (!new->masks_usage_stats) {
 		kfree(new);
 		return NULL;
 	}
 /* Flow lookup does full lookup on flow table. It starts with
  * mask from index passed in *index.
+ * This function MUST be called with BH disabled due to the use
+ * of CPU specific variables.
  */
 static struct sw_flow *flow_lookup(struct flow_table *tbl,
 				   struct table_instance *ti,
 				   u32 *n_cache_hit,
 				   u32 *index)
 {
-	u64 *usage_counters = this_cpu_ptr(ma->masks_usage_cntr);
+	struct mask_array_stats *stats = this_cpu_ptr(ma->masks_usage_stats);
 	struct sw_flow *flow;
 	struct sw_flow_mask *mask;
 	int i;
 		if (mask) {
 			flow = masked_flow_lookup(ti, key, mask, n_mask_hit);
 			if (flow) {
-				u64_stats_update_begin(&ma->syncp);
+				u64_stats_update_begin(&stats->syncp);
-				usage_counters[*index]++;
+				stats->usage_cntrs[*index]++;
-				u64_stats_update_end(&ma->syncp);
+				u64_stats_update_end(&stats->syncp);
 				(*n_cache_hit)++;
 				return flow;
 			}
 		flow = masked_flow_lookup(ti, key, mask, n_mask_hit);
 		if (flow) { /* Found */
 			*index = i;
-			u64_stats_update_begin(&ma->syncp);
+			u64_stats_update_begin(&stats->syncp);
-			usage_counters[*index]++;
+			stats->usage_cntrs[*index]++;
-			u64_stats_update_end(&ma->syncp);
+			u64_stats_update_end(&stats->syncp);
 			return flow;
 		}
 	}
 	struct mask_array *ma = rcu_dereference_ovsl(tbl->mask_array);
 	u32 __always_unused n_mask_hit;
 	u32 __always_unused n_cache_hit;
+	struct sw_flow *flow;
 	u32 index = 0;
-	return flow_lookup(tbl, ti, ma, key, &n_mask_hit, &n_cache_hit, &index);
+	/* This function gets called trough the netlink interface and therefore
+	 * is preemptible. However, flow_lookup() function needs to be called
+	 * with BH disabled due to CPU specific variables.
+	 */
+	local_bh_disable();
+	flow = flow_lookup(tbl, ti, ma, key, &n_mask_hit, &n_cache_hit, &index);
+	local_bh_enable();
+	return flow;
 }
 struct sw_flow *ovs_flow_tbl_lookup_exact(struct flow_table *tbl,
 	for (i = 0; i < ma->max; i++) {
 		struct sw_flow_mask *mask;
-		unsigned int start;
 		int cpu;
 		mask = rcu_dereference_ovsl(ma->masks[i]);
 		masks_and_count[i].counter = 0;
 		for_each_possible_cpu(cpu) {
-			u64 *usage_counters = per_cpu_ptr(ma->masks_usage_cntr,
+			struct mask_array_stats *stats;
-							  cpu);
+			unsigned int start;
 			u64 counter;
+			stats = per_cpu_ptr(ma->masks_usage_stats, cpu);
 			do {
-				start = u64_stats_fetch_begin_irq(&ma->syncp);
+				start = u64_stats_fetch_begin_irq(&stats->syncp);
-				counter = usage_counters[i];
+				counter = stats->usage_cntrs[i];
-			} while (u64_stats_fetch_retry_irq(&ma->syncp, start));
+			} while (u64_stats_fetch_retry_irq(&stats->syncp,
+							   start));
 			masks_and_count[i].counter += counter;
 		}

net/openvswitch/flow_table.h+6−2

 	u64 counter;
 };
+struct mask_array_stats {
+	struct u64_stats_sync syncp;
+	u64 usage_cntrs[];
+};
 struct mask_array {
 	struct rcu_head rcu;
 	int count, max;
-	u64 __percpu *masks_usage_cntr;
+	struct mask_array_stats __percpu *masks_usage_stats;
 	u64 *masks_usage_zero_cntr;
-	struct u64_stats_sync syncp;
 	struct sw_flow_mask __rcu *masks[];
 };

net/sched/act_ct.c+2−2

@@ -156,11 +156,11 @@ tcf_ct_flow_table_add_action_nat_udp(const struct nf_conntrack_tuple *tuple, · 155 unmodified lines
156	__be16 target_dst = target.dst.u.udp.port;	156	__be16 target_dst = target.dst.u.udp.port;
157		157
158	if (target_src != tuple->src.u.udp.port)	158	if (target_src != tuple->src.u.udp.port)
159	tcf_ct_add_mangle_action(action, FLOW_ACT_MANGLE_HDR_TYPE_TCP,	159	tcf_ct_add_mangle_action(action, FLOW_ACT_MANGLE_HDR_TYPE_UDP,
160	offsetof(struct udphdr, source),	160	offsetof(struct udphdr, source),
161	0xFFFF, be16_to_cpu(target_src));	161	0xFFFF, be16_to_cpu(target_src));
162	if (target_dst != tuple->dst.u.udp.port)	162	if (target_dst != tuple->dst.u.udp.port)
163	tcf_ct_add_mangle_action(action, FLOW_ACT_MANGLE_HDR_TYPE_TCP,	163	tcf_ct_add_mangle_action(action, FLOW_ACT_MANGLE_HDR_TYPE_UDP,
164	offsetof(struct udphdr, dest),	164	offsetof(struct udphdr, dest),
165	0xFFFF, be16_to_cpu(target_dst));	165	0xFFFF, be16_to_cpu(target_dst));
166	}	166	}

net/sched/act_tunnel_key.c+1−1

@@ -459,7 +459,7 @@ static int tunnel_key_init(struct net net, struct nlattr nla, · 458 unmodified lines
459		459
460	metadata = __ipv6_tun_set_dst(&saddr, &daddr, tos, ttl, dst_port,	460	metadata = __ipv6_tun_set_dst(&saddr, &daddr, tos, ttl, dst_port,
461	0, flags,	461	0, flags,
462	key_id, 0);	462	key_id, opts_len);
463	} else {	463	} else {
464	NL_SET_ERR_MSG(extack, "Missing either ipv4 or ipv6 src and dst");	464	NL_SET_ERR_MSG(extack, "Missing either ipv4 or ipv6 src and dst");
465	ret = -EINVAL;	465	ret = -EINVAL;

net/sched/cls_api.c+1−1

@@ -3712,7 +3712,7 @@ int tc_setup_flow_action(struct flow_action *flow_action, · 3711 unmodified lines
3712	entry->gate.num_entries = tcf_gate_num_entries(act);	3712	entry->gate.num_entries = tcf_gate_num_entries(act);
3713	err = tcf_gate_get_entries(entry, act);	3713	err = tcf_gate_get_entries(entry, act);
3714	if (err)	3714	if (err)
3715	goto err_out;	3715	goto err_out_locked;
3716	} else {	3716	} else {
3717	err = -EOPNOTSUPP;	3717	err = -EOPNOTSUPP;
3718	goto err_out_locked;	3718	goto err_out_locked;

net/tipc/bcast.c+8−2

 {
 	struct tipc_bc_base *bb = tipc_bc_base(net);
 	int all_dests =  tipc_link_bc_peers(bb->link);
+	int max_win = tipc_link_max_win(bb->link);
+	int min_win = tipc_link_min_win(bb->link);
 	int i, mtu, prim;
 	bb->primary_bearer = INVALID_BEARER_ID;
 			continue;
 		mtu = tipc_bearer_mtu(net, i);
-		if (mtu < tipc_link_mtu(bb->link))
+		if (mtu < tipc_link_mtu(bb->link)) {
 			tipc_link_set_mtu(bb->link, mtu);
+			tipc_link_set_queue_limits(bb->link,
+						   min_win,
+						   max_win);
+		}
 		bb->bcast_support &= tipc_bearer_bcast_support(net, i);
 		if (bb->dests[i] < all_dests)
 			continue;
 	if (max_win > TIPC_MAX_LINK_WIN)
 		return -EINVAL;
 	tipc_bcast_lock(net);
-	tipc_link_set_queue_limits(l, BCLINK_WIN_MIN, max_win);
+	tipc_link_set_queue_limits(l, tipc_link_min_win(l), max_win);
 	tipc_bcast_unlock(net);
 	return 0;
 }

samples/bpf/sockex3_kern.c+4−4

@@ -44,17 +44,17 @@ static inline void parse_eth_proto(struct __sk_buff *skb, u32 proto) · 43 unmodified lines
44	switch (proto) {	44	switch (proto) {
45	case ETH_P_8021Q:	45	case ETH_P_8021Q:
46	case ETH_P_8021AD:	46	case ETH_P_8021AD:
47	bpf_tail_call_static(skb, &jmp_table, PARSE_VLAN);	47	bpf_tail_call(skb, &jmp_table, PARSE_VLAN);
48	break;	48	break;
49	case ETH_P_MPLS_UC:	49	case ETH_P_MPLS_UC:
50	case ETH_P_MPLS_MC:	50	case ETH_P_MPLS_MC:
51	bpf_tail_call_static(skb, &jmp_table, PARSE_MPLS);	51	bpf_tail_call(skb, &jmp_table, PARSE_MPLS);
52	break;	52	break;
53	case ETH_P_IP:	53	case ETH_P_IP:
54	bpf_tail_call_static(skb, &jmp_table, PARSE_IP);	54	bpf_tail_call(skb, &jmp_table, PARSE_IP);
55	break;	55	break;
56	case ETH_P_IPV6:	56	case ETH_P_IPV6:
57	bpf_tail_call_static(skb, &jmp_table, PARSE_IPV6);	57	bpf_tail_call(skb, &jmp_table, PARSE_IPV6);
58	break;	58	break;
59	}	59	}
60	}	60	}

scripts/bpf_helpers_doc.py+1−0

@@ -453,6 +453,7 @@ class PrinterHelpers(Printer): · 452 unmodified lines
453	'struct bpf_perf_event_data',	453	'struct bpf_perf_event_data',
454	'struct bpf_perf_event_value',	454	'struct bpf_perf_event_value',
455	'struct bpf_pidns_info',	455	'struct bpf_pidns_info',
		456	'struct bpf_redir_neigh',
456	'struct bpf_sk_lookup',	457	'struct bpf_sk_lookup',
457	'struct bpf_sock',	458	'struct bpf_sock',
458	'struct bpf_sock_addr',	459	'struct bpf_sock_addr',

tools/include/uapi/linux/bpf.h+18−4

  * 	Return
  * 		The id is returned or 0 in case the id could not be retrieved.
  *
- * long bpf_redirect_neigh(u32 ifindex, u64 flags)
+ * long bpf_redirect_neigh(u32 ifindex, struct bpf_redir_neigh *params, int plen, u64 flags)
  * 	Description
  * 		Redirect the packet to another net device of index *ifindex*
  * 		and fill in L2 addresses from neighboring subsystem. This helper
  * 		is somewhat similar to **bpf_redirect**\ (), except that it
  * 		populates L2 addresses as well, meaning, internally, the helper
- * 		performs a FIB lookup based on the skb's networking header to
+ * 		relies on the neighbor lookup for the L2 address of the nexthop.
- * 		get the address of the next hop and then relies on the neighbor
+ *
- * 		lookup for the L2 address of the nexthop.
+ * 		The helper will perform a FIB lookup based on the skb's
+ * 		networking header to get the address of the next hop, unless
+ * 		this is supplied by the caller in the *params* argument. The
+ * 		*plen* argument indicates the len of *params* and should be set
+ * 		to 0 if *params* is NULL.
  *
  * 		The *flags* argument is reserved and must be 0. The helper is
  * 		currently only supported for tc BPF program types, and enabled
 	__u8	dmac[6];     /* ETH_ALEN */
 };
+struct bpf_redir_neigh {
+	/* network family for lookup (AF_INET, AF_INET6) */
+	__u32 nh_family;
+	/* network address of nexthop; skips fib lookup to find gateway */
+	union {
+		__be32		ipv4_nh;
+		__u32		ipv6_nh[4];  /* in6_addr; network order */
+	};
+};
 enum bpf_task_fd_type {
 	BPF_FD_TYPE_RAW_TRACEPOINT,	/* tp name */
 	BPF_FD_TYPE_TRACEPOINT,		/* tp name */

tools/lib/bpf/bpf_helpers.h+2−0

 /*
  * Helper function to perform a tail call with a constant/immediate map slot.
  */
+#if __clang_major__ >= 8 && defined(__bpf__)
 static __always_inline void
 bpf_tail_call_static(void *ctx, const void *map, const __u32 slot)
 {
 		     :: [ctx]"r"(ctx), [map]"r"(map), [slot]"i"(slot)
 		     : "r0", "r1", "r2", "r3", "r4", "r5");
 }
+#endif
 /*
  * Helper structure used by eBPF C program

tools/testing/selftests/bpf/prog_tests/ksyms_btf.c+39−18

 #include <bpf/libbpf.h>
 #include <bpf/btf.h>
 #include "test_ksyms_btf.skel.h"
+#include "test_ksyms_btf_null_check.skel.h"
 static int duration;
-void test_ksyms_btf(void)
+static void test_basic(void)
 {
 	__u64 runqueues_addr, bpf_prog_active_addr;
 	__u32 this_rq_cpu;
 	int this_bpf_prog_active;
 	struct test_ksyms_btf *skel = NULL;
 	struct test_ksyms_btf__data *data;
-	struct btf *btf;
-	int percpu_datasec;
 	int err;
 	err = kallsyms_find("runqueues", &runqueues_addr);
 	if (CHECK(err == -ENOENT, "ksym_find", "symbol 'bpf_prog_active' not found\n"))
 		return;
-	btf = libbpf_find_kernel_btf();
-	if (CHECK(IS_ERR(btf), "btf_exists", "failed to load kernel BTF: %ld\n",
-		  PTR_ERR(btf)))
-		return;
-	percpu_datasec = btf__find_by_name_kind(btf, ".data..percpu",
-						BTF_KIND_DATASEC);
-	if (percpu_datasec < 0) {
-		printf("%s:SKIP:no PERCPU DATASEC in kernel btf\n",
-		       __func__);
-		test__skip();
-		goto cleanup;
-	}
 	skel = test_ksyms_btf__open_and_load();
 	if (CHECK(!skel, "skel_open", "failed to open and load skeleton\n"))
 		goto cleanup;
 	      data->out__bpf_prog_active);
 cleanup:
-	btf__free(btf);
 	test_ksyms_btf__destroy(skel);
 }
+static void test_null_check(void)
+{
+	struct test_ksyms_btf_null_check *skel;
+	skel = test_ksyms_btf_null_check__open_and_load();
+	CHECK(skel, "skel_open", "unexpected load of a prog missing null check\n");
+	test_ksyms_btf_null_check__destroy(skel);
+}
+void test_ksyms_btf(void)
+{
+	int percpu_datasec;
+	struct btf *btf;
+	btf = libbpf_find_kernel_btf();
+	if (CHECK(IS_ERR(btf), "btf_exists", "failed to load kernel BTF: %ld\n",
+		  PTR_ERR(btf)))
+		return;
+	percpu_datasec = btf__find_by_name_kind(btf, ".data..percpu",
+						BTF_KIND_DATASEC);
+	btf__free(btf);
+	if (percpu_datasec < 0) {
+		printf("%s:SKIP:no PERCPU DATASEC in kernel btf\n",
+		       __func__);
+		test__skip();
+		return;
+	}
+	if (test__start_subtest("basic"))
+		test_basic();
+	if (test__start_subtest("null_check"))
+		test_null_check();
+}

tools/testing/selftests/bpf/progs/test_ksyms_btf_null_check.c+31−0

@@ -0,0 +1,31 @@
	1	// SPDX-License-Identifier: GPL-2.0
	2	/* Copyright (c) 2020 Facebook */
	3
	4	#include "vmlinux.h"
	5
	6	#include <bpf/bpf_helpers.h>
	7
	8	extern const struct rq runqueues __ksym; /* struct type global var. */
	9	extern const int bpf_prog_active __ksym; /* int type global var. */
	10
	11	SEC("raw_tp/sys_enter")
	12	int handler(const void *ctx)
	13	{
	14	struct rq *rq;
	15	int *active;
	16	__u32 cpu;
	17
	18	cpu = bpf_get_smp_processor_id();
	19	rq = (struct rq *)bpf_per_cpu_ptr(&runqueues, cpu);
	20	active = (int *)bpf_per_cpu_ptr(&bpf_prog_active, cpu);
	21	if (active) {
	22	/* READ_ONCE */
	23	(volatile int )active;
	24	/* !rq has not been tested, so verifier should reject. */
	25	(volatile int )(&rq->cpu);
	26	}
	27
	28	return 0;
	29	}
	30
	31	char _license[] SEC("license") = "GPL";

tools/testing/selftests/bpf/progs/test_tc_neigh.c+3−2

 // SPDX-License-Identifier: GPL-2.0
+#include <stddef.h>
 #include <stdint.h>
 #include <stdbool.h>
 	if (bpf_skb_store_bytes(skb, 0, &zero, sizeof(zero), 0) < 0)
 		return TC_ACT_SHOT;
-	return bpf_redirect_neigh(get_dev_ifindex(dev_src), 0);
+	return bpf_redirect_neigh(get_dev_ifindex(dev_src), NULL, 0, 0);
 }
 SEC("src_ingress") int tc_src(struct __sk_buff *skb)
 	if (bpf_skb_store_bytes(skb, 0, &zero, sizeof(zero), 0) < 0)
 		return TC_ACT_SHOT;
-	return bpf_redirect_neigh(get_dev_ifindex(dev_dst), 0);
+	return bpf_redirect_neigh(get_dev_ifindex(dev_dst), NULL, 0, 0);
 }
 char __license[] SEC("license") = "GPL";

tools/testing/selftests/bpf/progs/test_tc_neigh_fib.c+155−0

@@ -0,0 +1,155 @@
	1	// SPDX-License-Identifier: GPL-2.0
	2	#include <stdint.h>
	3	#include <stdbool.h>
	4	#include <stddef.h>
	5
	6	#include <linux/bpf.h>
	7	#include <linux/stddef.h>
	8	#include <linux/pkt_cls.h>
	9	#include <linux/if_ether.h>
	10	#include <linux/in.h>
	11	#include <linux/ip.h>
	12	#include <linux/ipv6.h>
	13
	14	#include <bpf/bpf_helpers.h>
	15	#include <bpf/bpf_endian.h>
	16
	17	#ifndef ctx_ptr
	18	# define ctx_ptr(field) (void *)(long)(field)
	19	#endif
	20
	21	#define AF_INET 2
	22	#define AF_INET6 10
	23
	24	static __always_inline int fill_fib_params_v4(struct __sk_buff *skb,
	25	struct bpf_fib_lookup *fib_params)
	26	{
	27	void *data_end = ctx_ptr(skb->data_end);
	28	void *data = ctx_ptr(skb->data);
	29	struct iphdr *ip4h;
	30
	31	if (data + sizeof(struct ethhdr) > data_end)
	32	return -1;
	33
	34	ip4h = (struct iphdr *)(data + sizeof(struct ethhdr));
	35	if ((void *)(ip4h + 1) > data_end)
	36	return -1;
	37
	38	fib_params->family = AF_INET;
	39	fib_params->tos = ip4h->tos;
	40	fib_params->l4_protocol = ip4h->protocol;
	41	fib_params->sport = 0;
	42	fib_params->dport = 0;
	43	fib_params->tot_len = bpf_ntohs(ip4h->tot_len);
	44	fib_params->ipv4_src = ip4h->saddr;
	45	fib_params->ipv4_dst = ip4h->daddr;
	46
	47	return 0;
	48	}
	49
	50	static __always_inline int fill_fib_params_v6(struct __sk_buff *skb,
	51	struct bpf_fib_lookup *fib_params)
	52	{
	53	struct in6_addr src = (struct in6_addr )fib_params->ipv6_src;
	54	struct in6_addr dst = (struct in6_addr )fib_params->ipv6_dst;
	55	void *data_end = ctx_ptr(skb->data_end);
	56	void *data = ctx_ptr(skb->data);
	57	struct ipv6hdr *ip6h;
	58
	59	if (data + sizeof(struct ethhdr) > data_end)
	60	return -1;
	61
	62	ip6h = (struct ipv6hdr *)(data + sizeof(struct ethhdr));
	63	if ((void *)(ip6h + 1) > data_end)
	64	return -1;
	65
	66	fib_params->family = AF_INET6;
	67	fib_params->flowinfo = 0;
	68	fib_params->l4_protocol = ip6h->nexthdr;
	69	fib_params->sport = 0;
	70	fib_params->dport = 0;
	71	fib_params->tot_len = bpf_ntohs(ip6h->payload_len);
	72	*src = ip6h->saddr;
	73	*dst = ip6h->daddr;
	74
	75	return 0;
	76	}
	77
	78	SEC("chk_egress") int tc_chk(struct __sk_buff *skb)
	79	{
	80	void *data_end = ctx_ptr(skb->data_end);
	81	void *data = ctx_ptr(skb->data);
	82	__u32 *raw = data;
	83
	84	if (data + sizeof(struct ethhdr) > data_end)
	85	return TC_ACT_SHOT;
	86
	87	return !raw[0] && !raw[1] && !raw[2] ? TC_ACT_SHOT : TC_ACT_OK;
	88	}
	89
	90	static __always_inline int tc_redir(struct __sk_buff *skb)
	91	{
	92	struct bpf_fib_lookup fib_params = { .ifindex = skb->ingress_ifindex };
	93	__u8 zero[ETH_ALEN * 2];
	94	int ret = -1;
	95
	96	switch (skb->protocol) {
	97	case __bpf_constant_htons(ETH_P_IP):
	98	ret = fill_fib_params_v4(skb, &fib_params);
	99	break;
	100	case __bpf_constant_htons(ETH_P_IPV6):
	101	ret = fill_fib_params_v6(skb, &fib_params);
	102	break;
	103	}
	104
	105	if (ret)
	106	return TC_ACT_OK;
	107
	108	ret = bpf_fib_lookup(skb, &fib_params, sizeof(fib_params), 0);
	109	if (ret == BPF_FIB_LKUP_RET_NOT_FWDED \|\| ret < 0)
	110	return TC_ACT_OK;
	111
	112	__builtin_memset(&zero, 0, sizeof(zero));
	113	if (bpf_skb_store_bytes(skb, 0, &zero, sizeof(zero), 0) < 0)
	114	return TC_ACT_SHOT;
	115
	116	if (ret == BPF_FIB_LKUP_RET_NO_NEIGH) {
	117	struct bpf_redir_neigh nh_params = {};
	118
	119	nh_params.nh_family = fib_params.family;
	120	__builtin_memcpy(&nh_params.ipv6_nh, &fib_params.ipv6_dst,
	121	sizeof(nh_params.ipv6_nh));
	122
	123	return bpf_redirect_neigh(fib_params.ifindex, &nh_params,
	124	sizeof(nh_params), 0);
	125
	126	} else if (ret == BPF_FIB_LKUP_RET_SUCCESS) {
	127	void *data_end = ctx_ptr(skb->data_end);
	128	struct ethhdr *eth = ctx_ptr(skb->data);
	129
	130	if (eth + 1 > data_end)
	131	return TC_ACT_SHOT;
	132
	133	__builtin_memcpy(eth->h_dest, fib_params.dmac, ETH_ALEN);
	134	__builtin_memcpy(eth->h_source, fib_params.smac, ETH_ALEN);
	135
	136	return bpf_redirect(fib_params.ifindex, 0);
	137	}
	138
	139	return TC_ACT_SHOT;
	140	}
	141
	142	/* these are identical, but keep them separate for compatibility with the
	143	* section names expected by test_tc_redirect.sh
	144	*/
	145	SEC("dst_ingress") int tc_dst(struct __sk_buff *skb)
	146	{
	147	return tc_redir(skb);
	148	}
	149
	150	SEC("src_ingress") int tc_src(struct __sk_buff *skb)
	151	{
	152	return tc_redir(skb);
	153	}
	154
	155	char __license[] SEC("license") = "GPL";

tools/testing/selftests/bpf/test_tc_redirect.sh+15−3

 	{ echo >&2 "timeout is not available"; exit 1; }
 command -v ping >/dev/null 2>&1 || \
 	{ echo >&2 "ping is not available"; exit 1; }
-command -v ping6 >/dev/null 2>&1 || \
+if command -v ping6 >/dev/null 2>&1; then PING6=ping6; else PING6=ping; fi
-	{ echo >&2 "ping6 is not available"; exit 1; }
 command -v perl >/dev/null 2>&1 || \
 	{ echo >&2 "perl is not available"; exit 1; }
 command -v jq >/dev/null 2>&1 || \
 	echo -e "${TEST}: ${GREEN}PASS${NC}"
 	TEST="ICMPv6 connectivity test"
-	ip netns exec ${NS_SRC} ping6 $PING_ARG ${IP6_DST}
+	ip netns exec ${NS_SRC} $PING6 $PING_ARG ${IP6_DST}
 	if [ $? -ne 0 ]; then
 		echo -e "${TEST}: ${RED}FAIL${NC}"
 		exit 1
 netns_setup_bpf()
 {
 	local obj=$1
+	local use_forwarding=${2:-0}
 	ip netns exec ${NS_FWD} tc qdisc add dev veth_src_fwd clsact
 	ip netns exec ${NS_FWD} tc filter add dev veth_src_fwd ingress bpf da obj $obj sec src_ingress
 	ip netns exec ${NS_FWD} tc filter add dev veth_dst_fwd ingress bpf da obj $obj sec dst_ingress
 	ip netns exec ${NS_FWD} tc filter add dev veth_dst_fwd egress  bpf da obj $obj sec chk_egress
+	if [ "$use_forwarding" -eq "1" ]; then
+		# bpf_fib_lookup() checks if forwarding is enabled
+		ip netns exec ${NS_FWD} sysctl -w net.ipv4.ip_forward=1
+		ip netns exec ${NS_FWD} sysctl -w net.ipv6.conf.veth_dst_fwd.forwarding=1
+		ip netns exec ${NS_FWD} sysctl -w net.ipv6.conf.veth_src_fwd.forwarding=1
+		return 0
+	fi
 	veth_src=$(ip netns exec ${NS_FWD} cat /sys/class/net/veth_src_fwd/ifindex)
 	veth_dst=$(ip netns exec ${NS_FWD} cat /sys/class/net/veth_dst_fwd/ifindex)
 netns_test_connectivity
 netns_cleanup
 netns_setup
+netns_setup_bpf test_tc_neigh_fib.o 1
+netns_test_connectivity
+netns_cleanup
+netns_setup
 netns_setup_bpf test_tc_peer.o
 netns_test_connectivity

tools/testing/selftests/bpf/verifier/sock.c+25−0

@@ -631,3 +631,28 @@ · 630 unmodified lines
631	.prog_type = BPF_PROG_TYPE_SK_REUSEPORT,	631	.prog_type = BPF_PROG_TYPE_SK_REUSEPORT,
632	.result = ACCEPT,	632	.result = ACCEPT,
633	},	633	},
		634	{
		635	"mark null check on return value of bpf_skc_to helpers",
		636	.insns = {
		637	BPF_LDX_MEM(BPF_DW, BPF_REG_1, BPF_REG_1, offsetof(struct __sk_buff, sk)),
		638	BPF_JMP_IMM(BPF_JNE, BPF_REG_1, 0, 2),
		639	BPF_MOV64_IMM(BPF_REG_0, 0),
		640	BPF_EXIT_INSN(),
		641	BPF_MOV64_REG(BPF_REG_6, BPF_REG_1),
		642	BPF_EMIT_CALL(BPF_FUNC_skc_to_tcp_sock),
		643	BPF_MOV64_REG(BPF_REG_7, BPF_REG_0),
		644	BPF_MOV64_REG(BPF_REG_1, BPF_REG_6),
		645	BPF_EMIT_CALL(BPF_FUNC_skc_to_tcp_request_sock),
		646	BPF_MOV64_REG(BPF_REG_8, BPF_REG_0),
		647	BPF_JMP_IMM(BPF_JNE, BPF_REG_8, 0, 2),
		648	BPF_MOV64_IMM(BPF_REG_0, 0),
		649	BPF_EXIT_INSN(),
		650	BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_7, 0),
		651	BPF_EXIT_INSN(),
		652	},
		653	.prog_type = BPF_PROG_TYPE_SCHED_CLS,
		654	.result = REJECT,
		655	.errstr = "invalid mem access",
		656	.result_unpriv = REJECT,
		657	.errstr_unpriv = "unknown func",
		658	},

tools/testing/selftests/net/config+1−0

@@ -33,3 +33,4 @@ CONFIG_KALLSYMS=y · 32 unmodified lines
33	CONFIG_TRACEPOINTS=y	33	CONFIG_TRACEPOINTS=y
34	CONFIG_NET_DROP_MONITOR=m	34	CONFIG_NET_DROP_MONITOR=m
35	CONFIG_NETDEVSIM=m	35	CONFIG_NETDEVSIM=m
		36	CONFIG_NET_FOU=m

tools/testing/selftests/net/forwarding/vxlan_asymmetric.sh+10−0

 	bridge fdb add 00:00:5e:00:01:01 dev br1 self local vlan 10
 	bridge fdb add 00:00:5e:00:01:01 dev br1 self local vlan 20
+	sysctl_set net.ipv4.conf.all.rp_filter 0
+	sysctl_set net.ipv4.conf.vlan10-v.rp_filter 0
+	sysctl_set net.ipv4.conf.vlan20-v.rp_filter 0
 }
 switch_destroy()
 {
+	sysctl_restore net.ipv4.conf.all.rp_filter
 	bridge fdb del 00:00:5e:00:01:01 dev br1 self local vlan 20
 	bridge fdb del 00:00:5e:00:01:01 dev br1 self local vlan 10
 	bridge fdb add 00:00:5e:00:01:01 dev br1 self local vlan 10
 	bridge fdb add 00:00:5e:00:01:01 dev br1 self local vlan 20
+	sysctl_set net.ipv4.conf.all.rp_filter 0
+	sysctl_set net.ipv4.conf.vlan10-v.rp_filter 0
+	sysctl_set net.ipv4.conf.vlan20-v.rp_filter 0
 }
 export -f ns_switch_create

tools/testing/selftests/net/forwarding/vxlan_symmetric.sh+10−0

 	bridge fdb add 00:00:5e:00:01:01 dev br1 self local vlan 10
 	bridge fdb add 00:00:5e:00:01:01 dev br1 self local vlan 20
+	sysctl_set net.ipv4.conf.all.rp_filter 0
+	sysctl_set net.ipv4.conf.vlan10-v.rp_filter 0
+	sysctl_set net.ipv4.conf.vlan20-v.rp_filter 0
 }
 switch_destroy()
 {
+	sysctl_restore net.ipv4.conf.all.rp_filter
 	bridge fdb del 00:00:5e:00:01:01 dev br1 self local vlan 20
 	bridge fdb del 00:00:5e:00:01:01 dev br1 self local vlan 10
 	bridge fdb add 00:00:5e:00:01:01 dev br1 self local vlan 10
 	bridge fdb add 00:00:5e:00:01:01 dev br1 self local vlan 20
+	sysctl_set net.ipv4.conf.all.rp_filter 0
+	sysctl_set net.ipv4.conf.vlan10-v.rp_filter 0
+	sysctl_set net.ipv4.conf.vlan20-v.rp_filter 0
 }
 export -f ns_switch_create

tools/testing/selftests/net/mptcp/config+1−0

@@ -1,4 +1,5 @@
1	CONFIG_MPTCP=y	1	CONFIG_MPTCP=y
		2	CONFIG_IPV6=y
2	CONFIG_MPTCP_IPV6=y	3	CONFIG_MPTCP_IPV6=y
3	CONFIG_INET_DIAG=m	4	CONFIG_INET_DIAG=m
4	CONFIG_INET_MPTCP_DIAG=m	5	CONFIG_INET_MPTCP_DIAG=m

tools/testing/selftests/net/rtnetlink.sh+5−0

 		return $ksft_skip
 	fi
+	if ! /sbin/modprobe -q -n fou; then
+		echo "SKIP: module fou is not found"
+		return $ksft_skip
+	fi
+	/sbin/modprobe -q fou
 	ip -netns "$testns" fou add port 7777 ipproto 47 2>/dev/null
 	if [ $? -ne 0 ];then
 		echo "FAIL: can't add fou port 7777, skipping test"