Merge tag 'net-6.5-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

Pull networking fixes from Jakub Kicinski:
 "Including fixes from ipsec and netfilter.

  No known outstanding regressions.

  Fixes to fixes:

   - virtio-net: set queues after driver_ok, avoid a potential race
     added by recent fix

   - Revert "vlan: Fix VLAN 0 memory leak", it may lead to a warning
     when VLAN 0 is registered explicitly

   - nf_tables:
      - fix false-positive lockdep splat in recent fixes
      - don't fail inserts if duplicate has expired (fix test failures)
      - fix races between garbage collection and netns dismantle

  Current release - new code bugs:

   - mlx5: Fix mlx5_cmd_update_root_ft() error flow

  Previous releases - regressions:

   - phy: fix IRQ-based wake-on-lan over hibernate / power off

  Previous releases - always broken:

   - sock: fix misuse of sk_under_memory_pressure() preventing system
     from exiting global TCP memory pressure if a single cgroup is under
     pressure

   - fix the RTO timer retransmitting skb every 1ms if linear option is
     enabled

   - af_key: fix sadb_x_filter validation, amment netlink policy

   - ipsec: fix slab-use-after-free in decode_session6()

   - macb: in ZynqMP resume always configure PS GTR for non-wakeup
     source

  Misc:

   - netfilter: set default timeout to 3 secs for sctp shutdown send and
     recv state (from 300ms), align with protocol timers"

* tag 'net-6.5-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (49 commits)
  ice: Block switchdev mode when ADQ is active and vice versa
  qede: fix firmware halt over suspend and resume
  net: do not allow gso_size to be set to GSO_BY_FRAGS
  sock: Fix misuse of sk_under_memory_pressure()
  sfc: don't fail probe if MAE/TC setup fails
  sfc: don't unregister flow_indr if it was never registered
  net: dsa: mv88e6xxx: Wait for EEPROM done before HW reset
  net/mlx5: Fix mlx5_cmd_update_root_ft() error flow
  net/mlx5e: XDP, Fix fifo overrun on XDP_REDIRECT
  i40e: fix misleading debug logs
  iavf: fix FDIR rule fields masks validation
  ipv6: fix indentation of a config attribute
  mailmap: add entries for Simon Horman
  broadcom: b44: Use b44_writephy() return value
  net: openvswitch: reject negative ifindex
  team: Fix incorrect deletion of ETH_P_8021AD protocol vid from slaves
  net: phy: broadcom: stub c45 read/write for 54810
  netfilter: nft_dynset: disallow object maps
  netfilter: nf_tables: GC transaction race with netns dismantle
  netfilter: nf_tables: fix GC transaction races with netns and netlink event exit path
  ...

Author: torvalds

.mailmap

@@ -538,6 +538,8 @@ Shuah Khan <shuah@kernel.org> <shuah.kh@samsung.com>
 Sibi Sankar <quic_sibis@quicinc.com> <sibis@codeaurora.org>
 Sid Manning <quic_sidneym@quicinc.com> <sidneym@codeaurora.org>
 Simon Arlott <simon@octiron.net> <simon@fire.lp0.eu>
+Simon Horman <horms@kernel.org> <simon.horman@corigine.com>
+Simon Horman <horms@kernel.org> <simon.horman@netronome.com>
 Simon Kelley <simon@thekelleys.org.uk>
 Sricharan Ramabadhran <quic_srichara@quicinc.com> <sricharan@codeaurora.org>
 Srinivas Ramana <quic_sramana@quicinc.com> <sramana@codeaurora.org>

Documentation/networking/nf_conntrack-sysctl.rst

@@ -178,10 +178,10 @@ nf_conntrack_sctp_timeout_established - INTEGER (seconds)
 	Default is set to (hb_interval * path_max_retrans + rto_max)
 
 nf_conntrack_sctp_timeout_shutdown_sent - INTEGER (seconds)
-	default 0.3
+	default 3
 
 nf_conntrack_sctp_timeout_shutdown_recd - INTEGER (seconds)
-	default 0.3
+	default 3
 
 nf_conntrack_sctp_timeout_shutdown_ack_sent - INTEGER (seconds)
 	default 3

drivers/net/dsa/mv88e6xxx/chip.c

@@ -3034,6 +3034,14 @@ static void mv88e6xxx_hardware_reset(struct mv88e6xxx_chip *chip)
 
 	/* If there is a GPIO connected to the reset pin, toggle it */
 	if (gpiod) {
+		/* If the switch has just been reset and not yet completed
+		 * loading EEPROM, the reset may interrupt the I2C transaction
+		 * mid-byte, causing the first EEPROM read after the reset
+		 * from the wrong location resulting in the switch booting
+		 * to wrong mode and inoperable.
+		 */
+		mv88e6xxx_g1_wait_eeprom_done(chip);
+
 		gpiod_set_value_cansleep(gpiod, 1);
 		usleep_range(10000, 20000);
 		gpiod_set_value_cansleep(gpiod, 0);

drivers/net/ethernet/broadcom/b44.c

@@ -1793,11 +1793,9 @@ static int b44_nway_reset(struct net_device *dev)
 	b44_readphy(bp, MII_BMCR, &bmcr);
 	b44_readphy(bp, MII_BMCR, &bmcr);
 	r = -EINVAL;
-	if (bmcr & BMCR_ANENABLE) {
-		b44_writephy(bp, MII_BMCR,
-			     bmcr | BMCR_ANRESTART);
-		r = 0;
-	}
+	if (bmcr & BMCR_ANENABLE)
+		r = b44_writephy(bp, MII_BMCR,
+				 bmcr | BMCR_ANRESTART);
 	spin_unlock_irq(&bp->lock);
 
 	return r;

drivers/net/ethernet/cadence/macb_main.c

@@ -5194,6 +5194,9 @@ static int __maybe_unused macb_suspend(struct device *dev)
 	unsigned int q;
 	int err;
 
+	if (!device_may_wakeup(&bp->dev->dev))
+		phy_exit(bp->sgmii_phy);
+
 	if (!netif_running(netdev))
 		return 0;
 
@@ -5254,7 +5257,6 @@ static int __maybe_unused macb_suspend(struct device *dev)
 	if (!(bp->wol & MACB_WOL_ENABLED)) {
 		rtnl_lock();
 		phylink_stop(bp->phylink);
-		phy_exit(bp->sgmii_phy);
 		rtnl_unlock();
 		spin_lock_irqsave(&bp->lock, flags);
 		macb_reset_hw(bp);
@@ -5284,6 +5286,9 @@ static int __maybe_unused macb_resume(struct device *dev)
 	unsigned int q;
 	int err;
 
+	if (!device_may_wakeup(&bp->dev->dev))
+		phy_init(bp->sgmii_phy);
+
 	if (!netif_running(netdev))
 		return 0;
 
@@ -5344,8 +5349,6 @@ static int __maybe_unused macb_resume(struct device *dev)
 	macb_set_rx_mode(netdev);
 	macb_restore_features(bp);
 	rtnl_lock();
-	if (!device_may_wakeup(&bp->dev->dev))
-		phy_init(bp->sgmii_phy);
 
 	phylink_start(bp->phylink);
 	rtnl_unlock();

drivers/net/ethernet/intel/i40e/i40e_nvm.c

@@ -210,11 +210,11 @@ static int i40e_read_nvm_word_srctl(struct i40e_hw *hw, u16 offset,
  * @hw: pointer to the HW structure.
  * @module_pointer: module pointer location in words from the NVM beginning
  * @offset: offset in words from module start
- * @words: number of words to write
- * @data: buffer with words to write to the Shadow RAM
+ * @words: number of words to read
+ * @data: buffer with words to read to the Shadow RAM
  * @last_command: tells the AdminQ that this is the last command
  *
- * Writes a 16 bit words buffer to the Shadow RAM using the admin command.
+ * Reads a 16 bit words buffer to the Shadow RAM using the admin command.
  **/
 static int i40e_read_nvm_aq(struct i40e_hw *hw,
 			    u8 module_pointer, u32 offset,
@@ -234,18 +234,18 @@ static int i40e_read_nvm_aq(struct i40e_hw *hw,
 	 */
 	if ((offset + words) > hw->nvm.sr_size)
 		i40e_debug(hw, I40E_DEBUG_NVM,
-			   "NVM write error: offset %d beyond Shadow RAM limit %d\n",
+			   "NVM read error: offset %d beyond Shadow RAM limit %d\n",
 			   (offset + words), hw->nvm.sr_size);
 	else if (words > I40E_SR_SECTOR_SIZE_IN_WORDS)
-		/* We can write only up to 4KB (one sector), in one AQ write */
+		/* We can read only up to 4KB (one sector), in one AQ write */
 		i40e_debug(hw, I40E_DEBUG_NVM,
-			   "NVM write fail error: tried to write %d words, limit is %d.\n",
+			   "NVM read fail error: tried to read %d words, limit is %d.\n",
 			   words, I40E_SR_SECTOR_SIZE_IN_WORDS);
 	else if (((offset + (words - 1)) / I40E_SR_SECTOR_SIZE_IN_WORDS)
 		 != (offset / I40E_SR_SECTOR_SIZE_IN_WORDS))
-		/* A single write cannot spread over two sectors */
+		/* A single read cannot spread over two sectors */
 		i40e_debug(hw, I40E_DEBUG_NVM,
-			   "NVM write error: cannot spread over two sectors in a single write offset=%d words=%d\n",
+			   "NVM read error: cannot spread over two sectors in a single read offset=%d words=%d\n",
 			   offset, words);
 	else
 		ret_code = i40e_aq_read_nvm(hw, module_pointer,

drivers/net/ethernet/intel/iavf/iavf_ethtool.c

@@ -1289,6 +1289,7 @@ iavf_add_fdir_fltr_info(struct iavf_adapter *adapter, struct ethtool_rx_flow_spe
 		fltr->ip_mask.src_port = fsp->m_u.tcp_ip4_spec.psrc;
 		fltr->ip_mask.dst_port = fsp->m_u.tcp_ip4_spec.pdst;
 		fltr->ip_mask.tos = fsp->m_u.tcp_ip4_spec.tos;
+		fltr->ip_ver = 4;
 		break;
 	case AH_V4_FLOW:
 	case ESP_V4_FLOW:
@@ -1300,6 +1301,7 @@ iavf_add_fdir_fltr_info(struct iavf_adapter *adapter, struct ethtool_rx_flow_spe
 		fltr->ip_mask.v4_addrs.dst_ip = fsp->m_u.ah_ip4_spec.ip4dst;
 		fltr->ip_mask.spi = fsp->m_u.ah_ip4_spec.spi;
 		fltr->ip_mask.tos = fsp->m_u.ah_ip4_spec.tos;
+		fltr->ip_ver = 4;
 		break;
 	case IPV4_USER_FLOW:
 		fltr->ip_data.v4_addrs.src_ip = fsp->h_u.usr_ip4_spec.ip4src;
@@ -1312,6 +1314,7 @@ iavf_add_fdir_fltr_info(struct iavf_adapter *adapter, struct ethtool_rx_flow_spe
 		fltr->ip_mask.l4_header = fsp->m_u.usr_ip4_spec.l4_4_bytes;
 		fltr->ip_mask.tos = fsp->m_u.usr_ip4_spec.tos;
 		fltr->ip_mask.proto = fsp->m_u.usr_ip4_spec.proto;
+		fltr->ip_ver = 4;
 		break;
 	case TCP_V6_FLOW:
 	case UDP_V6_FLOW:
@@ -1330,6 +1333,7 @@ iavf_add_fdir_fltr_info(struct iavf_adapter *adapter, struct ethtool_rx_flow_spe
 		fltr->ip_mask.src_port = fsp->m_u.tcp_ip6_spec.psrc;
 		fltr->ip_mask.dst_port = fsp->m_u.tcp_ip6_spec.pdst;
 		fltr->ip_mask.tclass = fsp->m_u.tcp_ip6_spec.tclass;
+		fltr->ip_ver = 6;
 		break;
 	case AH_V6_FLOW:
 	case ESP_V6_FLOW:
@@ -1345,6 +1349,7 @@ iavf_add_fdir_fltr_info(struct iavf_adapter *adapter, struct ethtool_rx_flow_spe
 		       sizeof(struct in6_addr));
 		fltr->ip_mask.spi = fsp->m_u.ah_ip6_spec.spi;
 		fltr->ip_mask.tclass = fsp->m_u.ah_ip6_spec.tclass;
+		fltr->ip_ver = 6;
 		break;
 	case IPV6_USER_FLOW:
 		memcpy(&fltr->ip_data.v6_addrs.src_ip, fsp->h_u.usr_ip6_spec.ip6src,
@@ -1361,6 +1366,7 @@ iavf_add_fdir_fltr_info(struct iavf_adapter *adapter, struct ethtool_rx_flow_spe
 		fltr->ip_mask.l4_header = fsp->m_u.usr_ip6_spec.l4_4_bytes;
 		fltr->ip_mask.tclass = fsp->m_u.usr_ip6_spec.tclass;
 		fltr->ip_mask.proto = fsp->m_u.usr_ip6_spec.l4_proto;
+		fltr->ip_ver = 6;
 		break;
 	case ETHER_FLOW:
 		fltr->eth_data.etype = fsp->h_u.ether_spec.h_proto;
@@ -1371,6 +1377,10 @@ iavf_add_fdir_fltr_info(struct iavf_adapter *adapter, struct ethtool_rx_flow_spe
 		return -EINVAL;
 	}
 
+	err = iavf_validate_fdir_fltr_masks(adapter, fltr);
+	if (err)
+		return err;
+
 	if (iavf_fdir_is_dup_fltr(adapter, fltr))
 		return -EEXIST;
 

drivers/net/ethernet/intel/iavf/iavf_fdir.c

@@ -18,6 +18,79 @@ static const struct in6_addr ipv6_addr_full_mask = {
 	}
 };
 
+static const struct in6_addr ipv6_addr_zero_mask = {
+	.in6_u = {
+		.u6_addr8 = {
+			0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+		}
+	}
+};
+
+/**
+ * iavf_validate_fdir_fltr_masks - validate Flow Director filter fields masks
+ * @adapter: pointer to the VF adapter structure
+ * @fltr: Flow Director filter data structure
+ *
+ * Returns 0 if all masks of packet fields are either full or empty. Returns
+ * error on at least one partial mask.
+ */
+int iavf_validate_fdir_fltr_masks(struct iavf_adapter *adapter,
+				  struct iavf_fdir_fltr *fltr)
+{
+	if (fltr->eth_mask.etype && fltr->eth_mask.etype != htons(U16_MAX))
+		goto partial_mask;
+
+	if (fltr->ip_ver == 4) {
+		if (fltr->ip_mask.v4_addrs.src_ip &&
+		    fltr->ip_mask.v4_addrs.src_ip != htonl(U32_MAX))
+			goto partial_mask;
+
+		if (fltr->ip_mask.v4_addrs.dst_ip &&
+		    fltr->ip_mask.v4_addrs.dst_ip != htonl(U32_MAX))
+			goto partial_mask;
+
+		if (fltr->ip_mask.tos && fltr->ip_mask.tos != U8_MAX)
+			goto partial_mask;
+	} else if (fltr->ip_ver == 6) {
+		if (memcmp(&fltr->ip_mask.v6_addrs.src_ip, &ipv6_addr_zero_mask,
+			   sizeof(struct in6_addr)) &&
+		    memcmp(&fltr->ip_mask.v6_addrs.src_ip, &ipv6_addr_full_mask,
+			   sizeof(struct in6_addr)))
+			goto partial_mask;
+
+		if (memcmp(&fltr->ip_mask.v6_addrs.dst_ip, &ipv6_addr_zero_mask,
+			   sizeof(struct in6_addr)) &&
+		    memcmp(&fltr->ip_mask.v6_addrs.dst_ip, &ipv6_addr_full_mask,
+			   sizeof(struct in6_addr)))
+			goto partial_mask;
+
+		if (fltr->ip_mask.tclass && fltr->ip_mask.tclass != U8_MAX)
+			goto partial_mask;
+	}
+
+	if (fltr->ip_mask.proto && fltr->ip_mask.proto != U8_MAX)
+		goto partial_mask;
+
+	if (fltr->ip_mask.src_port && fltr->ip_mask.src_port != htons(U16_MAX))
+		goto partial_mask;
+
+	if (fltr->ip_mask.dst_port && fltr->ip_mask.dst_port != htons(U16_MAX))
+		goto partial_mask;
+
+	if (fltr->ip_mask.spi && fltr->ip_mask.spi != htonl(U32_MAX))
+		goto partial_mask;
+
+	if (fltr->ip_mask.l4_header &&
+	    fltr->ip_mask.l4_header != htonl(U32_MAX))
+		goto partial_mask;
+
+	return 0;
+
+partial_mask:
+	dev_err(&adapter->pdev->dev, "Failed to add Flow Director filter, partial masks are not supported\n");
+	return -EOPNOTSUPP;
+}
+
 /**
  * iavf_pkt_udp_no_pay_len - the length of UDP packet without payload
  * @fltr: Flow Director filter data structure
@@ -263,8 +336,6 @@ iavf_fill_fdir_ip4_hdr(struct iavf_fdir_fltr *fltr,
 		VIRTCHNL_ADD_PROTO_HDR_FIELD_BIT(hdr, IPV4, DST);
 	}
 
-	fltr->ip_ver = 4;
-
 	return 0;
 }
 
@@ -309,8 +380,6 @@ iavf_fill_fdir_ip6_hdr(struct iavf_fdir_fltr *fltr,
 		VIRTCHNL_ADD_PROTO_HDR_FIELD_BIT(hdr, IPV6, DST);
 	}
 
-	fltr->ip_ver = 6;
-
 	return 0;
 }
 

drivers/net/ethernet/intel/iavf/iavf_fdir.h

@@ -110,6 +110,8 @@ struct iavf_fdir_fltr {
 	struct virtchnl_fdir_add vc_add_msg;
 };
 
+int iavf_validate_fdir_fltr_masks(struct iavf_adapter *adapter,
+				  struct iavf_fdir_fltr *fltr);
 int iavf_fill_fdir_add_msg(struct iavf_adapter *adapter, struct iavf_fdir_fltr *fltr);
 void iavf_print_fdir_fltr(struct iavf_adapter *adapter, struct iavf_fdir_fltr *fltr);
 bool iavf_fdir_is_dup_fltr(struct iavf_adapter *adapter, struct iavf_fdir_fltr *fltr);

drivers/net/ethernet/intel/ice/ice_eswitch.c

@@ -538,6 +538,12 @@ ice_eswitch_mode_set(struct devlink *devlink, u16 mode,
 		break;
 	case DEVLINK_ESWITCH_MODE_SWITCHDEV:
 	{
+		if (ice_is_adq_active(pf)) {
+			dev_err(ice_pf_to_dev(pf), "Couldn't change eswitch mode to switchdev - ADQ is active. Delete ADQ configs and try again, e.g. tc qdisc del dev $PF root");
+			NL_SET_ERR_MSG_MOD(extack, "Couldn't change eswitch mode to switchdev - ADQ is active. Delete ADQ configs and try again, e.g. tc qdisc del dev $PF root");
+			return -EOPNOTSUPP;
+		}
+
 		dev_info(ice_pf_to_dev(pf), "PF %d changed eswitch mode to switchdev",
 			 pf->hw.pf_id);
 		NL_SET_ERR_MSG_MOD(extack, "Changed eswitch mode to switchdev");